CN119804330B

CN119804330B - Self-adaptive regulation and control method and device for wafer detection light source brightness

Info

Publication number: CN119804330B
Application number: CN202510301440.9A
Authority: CN
Inventors: 王敏军; 李国鑫
Original assignee: Sundopt Led Lighting Co ltd
Current assignee: Sundopt Led Lighting Co ltd
Priority date: 2025-03-14
Filing date: 2025-03-14
Publication date: 2025-06-13
Anticipated expiration: 2045-03-14
Also published as: CN119804330A

Abstract

The invention relates to the technical field of light source control, and discloses a self-adaptive regulation and control method and device for wafer detection light source brightness, the invention performs preliminary irradiation on a wafer through a preset annular light source, acquires initial image data, performs digital simulation according to the initial detection data, generates a wafer detection feedback model, analyzes the light source irradiation requirement according to the detection feedback model, generates a gradient execution scheme, sequentially configures light source parameters according to the gradient scheme, performs multi-round detection on the wafer and collects data, the method realizes high-efficiency and accurate wafer surface defect detection by automatically adjusting the brightness and configuration of the light source, improves the detection precision and efficiency, adapts to the detection requirements of different wafers, and solves the problem that the defects are not detected due to insufficient comprehensive light sources for detection in the prior art.

Description

Self-adaptive regulation and control method and device for wafer detection light source brightness

Technical Field

The invention relates to the technical field of light source control, in particular to a self-adaptive regulation and control method and device for detecting the brightness of a light source by a wafer.

Background

The wafer is used as a core material in semiconductor manufacturing, the surface quality of the wafer directly influences the performance and the reliability of chips, various defects including micro cracks, scratches, pollutants and the like possibly exist on the surface of the wafer, the defects are difficult to find by a traditional detection means, a light source is one of the most critical elements in the detection of the surface of the wafer, different reflection characteristics can be generated when different types of light sources (such as a ring light source, a laser light source, a polarized light source and the like) irradiate the surface of the wafer, so that different defect information is disclosed, and parameters such as brightness, angle and distribution of the light source need to be adjusted according to the actual state of the wafer so as to realize the optimal detection effect.

In the actual wafer inspection process, the adjustment of the brightness of the light source is usually performed empirically, which may cause some potential defects not to be detected or cause the image to be overexposed due to too strong reflection, and the traditional method cannot automatically adjust the brightness of the light source according to the real-time condition of the reflection of the wafer surface, so that the inspection requirements of different types of defects are difficult to deal with.

Disclosure of Invention

The invention aims to provide a self-adaptive regulation and control method and device for brightness of a wafer detection light source, and aims to solve the problem that defects are not detected due to insufficient comprehensive light sources for detection in the prior art.

The invention is realized in such a way, and in a first aspect, the invention provides a self-adaptive regulation and control method for detecting the brightness of a light source by a wafer, which comprises the following steps:

The method comprises the steps that initial light source irradiation in a surrounding state is conducted on a wafer to be detected through a preset annular light source, and image data of the wafer to be detected under the irradiation of the initial light source in the surrounding state are collected to obtain initial detection data of the wafer to be detected;

Performing digital simulation on the wafer to be detected according to the initial detection data of the wafer to be detected to obtain a wafer detection feedback model, and performing gradient scheme analysis on the follow-up light source irradiation requirement of the annular light source based on the wafer detection feedback model to obtain a light source irradiation gradient execution scheme;

sequentially configuring light source parameters of the annular light source according to the light source irradiation gradient execution scheme, so that the annular light source sequentially applies detection light source irradiation to the wafer to be detected, and gradient detection data of the wafer to be detected under the irradiation of each round of detection light source are obtained;

Substituting gradient detection data of the wafer to be detected under each round of detection light source into the wafer detection feedback model so as to carry out model parameter adjustment on the wafer detection feedback model;

and carrying out expansion light source detection demand analysis according to the wafer detection feedback model subjected to model parameter adjustment so as to obtain an expansion light source detection scheme of the wafer to be detected, detecting the wafer to be detected according to the expansion light source detection scheme, obtaining expansion detection data and substituting the expansion detection data into the wafer detection feedback model.

In a second aspect, the present invention provides an adaptive adjustment and control device for brightness of a wafer inspection light source, which is configured to implement the adaptive adjustment and control method for brightness of a wafer inspection light source according to any one of the first aspect, including:

the initial detection module is used for carrying out initial light source irradiation of a surrounding state on a wafer to be detected through a preset annular light source, and collecting image data of the wafer to be detected under the irradiation of the surrounding initial light source so as to obtain initial detection data of the wafer to be detected;

The scheme analysis module is used for carrying out digital simulation on the wafer to be detected according to the initial detection data of the wafer to be detected to obtain the wafer detection feedback model, and carrying out gradient scheme analysis on the follow-up light source irradiation requirement of the annular light source based on the wafer detection feedback model to obtain a light source irradiation gradient execution scheme;

The gradient detection module is used for sequentially configuring light source parameters of the annular light source according to the light source irradiation gradient execution scheme, so that the annular light source sequentially applies detection light source irradiation to the wafer to be detected, and gradient detection data of the wafer to be detected under the irradiation of each round of detection light source are obtained;

the data feedback module is used for substituting the gradient detection data of the wafer to be detected under each round of detection light source into the wafer detection feedback model so as to carry out model parameter adjustment on the wafer detection feedback model;

The expansion detection module is used for carrying out expansion light source detection demand analysis according to the wafer detection feedback model subjected to model parameter adjustment so as to obtain an expansion light source detection scheme of the wafer to be detected, detecting the wafer to be detected according to the expansion light source detection scheme, obtaining expansion detection data and substituting the expansion detection data into the wafer detection feedback model.

The invention provides a self-adaptive regulation and control method for the brightness of a wafer detection light source, which has the following beneficial effects:

According to the method, initial irradiation is carried out on the wafer through the preset annular light source, initial image data are collected, digital simulation is carried out according to the initial detection data, a wafer detection feedback model is generated, a light source irradiation requirement is analyzed according to the detection feedback model, a gradient execution scheme is generated, light source parameters are sequentially configured according to the gradient scheme, multiple rounds of detection are carried out on the wafer, data are collected, the detection data are fed back to the model, parameter adjustment is carried out, the adjusted model carries out light source detection requirement analysis, an expanded detection scheme is generated, final detection is carried out, and through automatic adjustment of light source brightness and configuration, efficient and accurate detection of the surface defects of the wafer is achieved, detection precision and efficiency are improved, detection requirements of different wafers are met, and the problem that defects are not detected due to the fact that light sources used for detection are not comprehensive in the prior art is solved.

Drawings

Fig. 1 is a schematic diagram of steps of a method for adaptively adjusting brightness of a wafer inspection light source according to an embodiment of the present invention;

Fig. 2 is a schematic structural diagram of an adaptive adjustment device for detecting brightness of a light source for a wafer according to an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

The implementation of the present invention will be described in detail below with reference to specific embodiments.

Referring to fig. 1 and 2, a preferred embodiment of the present invention is provided.

In a first aspect, the present invention provides a method for adaptively adjusting and controlling brightness of a wafer inspection light source, including:

S1, carrying out initial light source irradiation of a surrounding state on a wafer to be detected through a preset annular light source, and collecting image data of the wafer to be detected under the irradiation of the surrounding initial light source to obtain initial detection data of the wafer to be detected;

S2, performing digital simulation on the wafer to be detected according to the initial detection data of the wafer to be detected to obtain a wafer detection feedback model, and performing gradient scheme analysis on the follow-up light source irradiation requirement of the annular light source based on the wafer detection feedback model to obtain a light source irradiation gradient execution scheme;

s3, sequentially configuring light source parameters of the annular light source according to the light source irradiation gradient execution scheme, so that the annular light source sequentially irradiates the wafer to be detected with detection light sources, and gradient detection data of the wafer to be detected under the irradiation of all rounds of detection light sources are obtained;

S4, substituting gradient detection data of the wafer to be detected under each round of detection light sources into the wafer detection feedback model so as to carry out model parameter adjustment on the wafer detection feedback model;

S5, performing expansion light source detection demand analysis according to the wafer detection feedback model subjected to model parameter adjustment to obtain an expansion light source detection scheme of the wafer to be detected, detecting the wafer to be detected according to the expansion light source detection scheme to obtain expansion detection data, and substituting the expansion detection data into the wafer detection feedback model.

Specifically, in step S1 of the embodiment provided by the present invention, an annular light source is pre-installed and disposed, the light source generally surrounds the periphery of the wafer to form a uniform illumination environment, brightness, angle, position, etc. of the annular light source can be adjusted according to requirements to ensure that the entire surface of the wafer can be covered, the wafer to be detected is placed in the illumination range of the light source to ensure that the surface of the wafer can be uniformly illuminated by the annular light source, and when the annular light source illuminates the wafer, the annular light source adopts a manner of surrounding illumination, that is, the light sources are uniformly distributed along the periphery of the wafer to ensure that illumination intensity of each angle is basically consistent.

It can be appreciated that the configuration of the annular light source can provide uniform illumination, so that the problems of shadow and nonuniform illumination possibly generated by the conventional light source are avoided, the quality during image acquisition is improved, the comprehensiveness of image data is ensured, the annular light source can ensure that the surface of a wafer is uniformly illuminated at all angles, defect omission or misjudgment caused by insufficient local illumination is reduced, and compared with the conventional point light source, the annular light source can illuminate the surface of the wafer from multiple angles, so that the brightness uniformity of a primary image is improved, and the influence on detection is reduced.

More specifically, a high-resolution camera or optical detection device is used to ensure that the tiny structure and defect of the wafer surface can be accurately captured, image data of the wafer surface is collected under the irradiation of an annular light source to obtain detailed image information containing all areas of the wafer, the image data generally comprises the characteristics of surface textures, defects, flaws and the like of the wafer, the collected image data is transmitted to a processing system for subsequent image analysis and processing, the image collection process can obtain high-quality and high-contrast images due to the uniformity of the light source and the sufficient irradiation obtained on the wafer surface, the detection precision of the wafer defects is further improved, the irradiation of each corner of the wafer surface is ensured due to the design of the annular light source, the detection dead zone caused by irradiation dead angles is avoided, and the image data of all areas of the wafer surface are effectively collected.

More specifically, the collected image data is stored as an original image data file, usually in the form of an image file (such as TIFF, JPEG, PNG, etc.) or an image data set (such as matrix data), etc., which are used as basic data for subsequent detection and analysis, and then image processing and defect recognition are performed, and according to the collected image data, the image quality is evaluated, whether problems such as blurring and noise exist is checked to determine whether the light source parameters need to be adjusted or the data need to be collected again, the basic data is provided for subsequent defect detection and light source optimization by the image data collection at this stage, the collected initial data can be further optimized for the light source irradiation and detection method according to the initial data, and the collected initial data can be directly used for subsequent defect analysis and also provides powerful support for establishing a model of the wafer surface characteristics.

It can be understood that the technical effects of the whole steps are that the illumination condition is uniform during image acquisition by the uniform annular light source illumination, the influence of shadows and uneven illumination on the image is reduced to the greatest extent, the quality and reliability of detected images are further improved, the annular light source ensures that each area of a wafer can be fully illuminated, so that the possible illumination blind areas of the traditional light source are avoided, the acquisition of the initial image data is the basis for the regulation and control of all subsequent light sources and defect detection, and the accuracy and the high efficiency of the subsequent steps are ensured.

Specifically, in step S2 of the embodiment provided by the present invention, the image data collected by the preliminary annular light source irradiation is used as input for analysis by the digital simulation program, at this time, the image data may include information such as defects, textures, and material non-uniformity on the wafer surface, advanced simulation software or algorithm (such as finite element analysis, optical simulation, etc.) is used to model the wafer surface according to the input initial detection data, a virtual wafer surface model is generated, based on the wafer surface model, image responses under different light source irradiation conditions are simulated to obtain a wafer detection feedback model, the model displays the response characteristics of the wafer under different light conditions, including factors such as the visibility, brightness, contrast, etc., the virtual model generated by digital simulation can accurately reflect the real physical characteristics of the wafer, so that subsequent light source adjustment and defect detection are more accurate, simulate the detection feedback under different light source conditions, which regions of defects are difficult to be identified under the existing light conditions, thereby providing a basis for subsequent light source adjustment, the wafer detection feedback provides a theoretical optimal light source regulation and control parameter, and unnecessary light source optimization is avoided.

More specifically, based on the generated wafer detection feedback model, the irradiation requirements of different areas of the wafer surface are analyzed, for example, some areas may need stronger illumination due to the surface structure or the nature of defects, other areas may need weaker illumination to avoid overexposure or over illumination, a gradient scheme of light source irradiation is formulated according to the area requirements of the model feedback, the gradient scheme will specify parameters such as light source intensity and irradiation angle required by each area in detail, specifically, the required illumination intensity of different areas may be different, the illumination intensity of different areas may need to be optimized by adjusting the light source brightness, the irradiation angles of different areas may be finely adjusted according to the types of defects, for example, some defects are more obvious under oblique illumination, the influence of different gradient schemes on the wafer is simulated, the visibility and the detection effect of the defects of the wafer under each scheme are evaluated, and finally the optimal light source irradiation gradient execution scheme is determined.

It can be understood that through the accurate analysis and gradient adjustment of the illumination requirements of different areas of the wafer, the problem of excessive illumination or uneven illumination caused by overall unified light source illumination can be avoided, each area is ensured to obtain the most suitable illumination, certain small or weak defects are easier to find under specific illumination conditions, the detection sensitivity of the micro defects can be effectively improved through the adjustment of the gradient light source illumination scheme, the overexposure of some areas or the illumination shortage of other areas can be avoided through the flexible adjustment of the illumination intensity and angle of the light source, and the defect detection accuracy is improved to the greatest extent.

More specifically, according to the foregoing analysis of the light source gradient scheme, a specific light source irradiation gradient execution scheme is formed, the scheme should detail parameters such as light source intensity, angle, irradiation time and the like required by each region, and provide implementation details, the gradient execution scheme is applied in an actual detection environment, and the effect is verified through experiments, if some regions still have undetected defects, it may be necessary to further adjust the light source irradiation scheme, after several rounds of adjustment and optimization, the final gradient execution scheme is confirmed, so that the defects of each region can be accurately detected, false alarm and false alarm are avoided, and by means of the accurate gradient light source scheme, the detection of the wafer surface can be quickly and efficiently completed, meanwhile, the surface characteristics and defect types of different wafers can be adapted, by optimizing the light source irradiation mode, the detection accuracy of different defect types and different regions can be obviously improved, and all defects (whether size or type) can be effectively captured.

It can be understood that through analysis and adjustment of the gradient light source scheme, each wafer area can be properly irradiated according to actual requirements, so that the problem of insufficient or excessive irradiation caused by uniform light source irradiation in the traditional method is avoided, digital simulation and feedback model optimization under different illumination conditions can clearly show tiny and complex defects under the proper illumination conditions, the detection accuracy is improved, the method can dynamically adjust the light source parameters according to actual detection requirements, adapt to the detection requirements of different types of wafers and defects, has stronger flexibility and adaptability, reduces unnecessary detection steps through model simulation and gradient scheme analysis in the earlier stage, and finally improves the efficiency and accuracy of the whole detection process.

Specifically, in step S3 of the embodiment provided by the present invention, the irradiation parameters of the light source are gradually adjusted according to the determined gradient light source irradiation scheme, which mainly includes adjusting the light source intensity (brightness) according to the requirements of different areas to ensure that the wafer surfaces of the different areas are irradiated most appropriately, adjusting the angle between the light source and the wafer surfaces to ensure the illumination effect of the different areas, especially the appearance effect of surface defects, and adjusting the irradiation time according to the required exposure time to ensure that each area is irradiated sufficiently without overexposure or underexposure.

More specifically, the annular light source is adjusted sequentially according to a set gradient scheme, and irradiates the surface of the wafer sequentially, in the irradiation process of each round of light source, parameters of the light source can be changed according to the gradient scheme so as to ensure that each region obtains different irradiation conditions, after each round of light source is irradiated, a detection device (such as an optical imaging system or a camera) is used for capturing light signals reflected or transmitted by the surface of the wafer, image data or other detection data of the wafer under different irradiation conditions are obtained, the data reflect the response of the surface of the wafer under different light source irradiation conditions, the response comprises defect information, surface non-uniformity, reflectivity and the like, the light source parameters are adjusted gradually, and the requirements of each region are optimized, so that the surface of the wafer in each region can be irradiated and detected under the optimal conditions, more accurate detection data can be obtained, different defect types and different expression forms of the surface of the wafer can be covered through irradiation under the condition of different light sources, the comprehensive performance of defect detection can be increased, and the customized detection flow can be realized by self-adapting to the detection requirements of different wafers and different defect types through accurate adjustment of the angles and time.

More specifically, after each light source irradiation, the detection system (such as a CCD camera or an optical sensor) can collect the reflected or transmitted light signals on the surface of the wafer in real time, the signals can be converted into image data or other forms of feedback data, the collected image or light signal data is subjected to preliminary processing and calibration, noise is removed, the contrast, brightness and other key information of the image are enhanced, the defect can be clearly seen, the data after each round of irradiation can be recorded and stored for subsequent analysis and comparison, and parameters (such as light source intensity, angle and time) of each irradiation can be stored together with the corresponding detection data to form a complete gradient detection data set.

More specifically, through the data acquisition under a plurality of gradient illumination schemes, multi-dimensional detection data can be obtained, including reflection, transmission, contrast and other information under different illumination conditions. The data can provide omnibearing reference for subsequent analysis, key information on the surface of the wafer can be extracted to the greatest extent in each round of irradiation by gradually adjusting the parameters of the light source, the quality of defect identification is improved, and the data irradiated by different light sources can help to identify defects of different types, such as micro cracks, surface pollution, uneven areas and the like, so that the defects on the surface of the wafer can be more accurately classified and positioned.

More specifically, by comparing the data collected under different light source irradiation conditions, analyzing which irradiation schemes have more obvious defects on the surface of the wafer and which areas may have missed detection or false detection, further adjusting the light source irradiation gradient according to the data analysis result, improving the distribution of light source intensity, angle or time, and optimizing the detection effect. The detection flow is continuously improved through experiments and data feedback to achieve an optimal detection result, and the data collected under the irradiation of each round of gradient light source are utilized to identify and position defects by combining a defect identification algorithm (such as image processing, machine learning and the like), so that the defect characteristics of the surface of the wafer are further extracted.

More specifically, through the data comparison and analysis under different light source conditions, the method can help accurately identify different kinds of defects, especially tiny defects or complex defects which are difficult to find under the conventional light source, and through the feedback of the irradiation data of the gradient light source, the self-optimization and dynamic adjustment can be realized, so that the light source configuration can be more accurately adapted to the surface characteristics and defect types of different wafers, the overall detection capability is improved, the occurrence of missed detection and false detection is effectively reduced through the gradient irradiation round by round and the multiple data comparison, and the reliability and the accuracy of detection are improved.

It can be understood that through irradiation and data acquisition under the condition of multiple rounds of different light sources, the wafer surface can be covered completely, any potential defect area is not missed, all defects can be found, the annular light source is used for gradually adjusting the light source parameters according to a gradient execution scheme, the light source irradiation is precisely controlled, the unnecessary overexposure or insufficient irradiation problem is avoided, the accuracy of wafer defect detection is improved, the detection flow is continuously optimized through repeated light source adjustment and data feedback, the detection efficiency is improved, the optimal detection effect is ensured, and various defects, especially the detection of tiny or complex defects, can be more clearly and precisely identified based on gradient data under different light source irradiation conditions.

Specifically, in step S4 of the embodiment provided by the present invention, in the previous step, wafer surface detection data (such as reflectivity, transmission image, optical image, etc.) under different light source conditions are obtained by successively adjusting light source parameters and performing irradiation, and necessary preprocessing including noise removal, image enhancement (such as contrast adjustment, smoothing processing), normalization, etc. is performed on the obtained gradient detection data. The process ensures that the quality of data input into the model is high and consistent, the processed gradient detection data is input into the wafer detection feedback model to serve as input characteristics of the model, at the moment, the gradient data is not only original information of each round of detection, but also information related to environmental conditions and detection parameters under irradiation of each round of light source, the quality of the input data of the model is ensured by the preprocessed gradient detection data, and therefore the training and adjustment effects of the follow-up model are improved.

More specifically, the wafer inspection feedback model is typically initialized based on some preliminary model parameters (such as threshold, weight, bias, etc.), forming an initial inspection state, this process may be performed based on historical data or preliminary assumptions, model predictions are made using the input gradient inspection data, defect probability or defect type predictions for each region are obtained, model predictions and actual inspection results (if there is ground truth data) are compared, errors (such as deviation of predicted defect position from actual defect position, false detection rate, miss detection rate, etc.) are calculated, and based on the error values, parameters of the model are adjusted by a back propagation algorithm (in a deep learning model) or other optimization algorithm (such as gradient descent, genetic algorithm, etc.), and the update process typically includes adjusting model parameters of weight, bias, threshold, etc. to reduce the prediction errors.

More specifically, under the condition of multiple rounds of light sources, the model can continuously learn and self-adjust, the recognition accuracy of the model to the surface defects of the wafer is gradually improved through repeated iterative optimization, the adjusted model needs to be verified through new data, and the adjusted parameters can be confirmed to obviously improve the defect detection accuracy. If errors are found to still exist, iteratively adjusting the model parameters until an ideal detection effect is obtained, and the model can adaptively optimize the model parameters by substituting actual gradient detection data and combining error back propagation. Along with the adjustment of each round, the detectability of model can constantly promote, and the model passes through fine parameter adjustment, can discern the defect under the different illumination conditions more accurately, gradually improves the defect identification ability under the complex scene, along with the continuous input of gradient data, the model can constantly adapt to the changing factors such as different grade wafers, different defect types, different light source conditions for detecting system possesses stronger dynamic adaptability.

More specifically, after each model adjustment, new detection results can be used as data feedback again, a second round of even multiple rounds of learning can be performed, the model can be further optimized by inputting feedback data of each round, so that the model becomes more accurate, the generalization capability of the model is ensured not to be influenced by fitting through Cross-validation (Cross-effect) and other technologies, the process is beneficial to improving the robustness of the model under different wafers and different light source conditions, after multiple rounds of iteration and optimization, the final model can be comprehensively evaluated based on multidimensional data (including data under different light source irradiation conditions), the performance of the model in a real production environment is checked, such as defect detection rate, omission rate, false detection rate and the like, the identification accuracy of the model can be continuously improved through repeated iteration and parameter adjustment, finally high-accuracy defect detection can be realized under a complex process environment, the model can adapt to different process changes, wafer defects of different types and sizes can be processed, the stability of the model can be kept in the continuous changed production environment and the continuous process changes can be better, the overall detection accuracy can be improved along with the improvement of the quality detection accuracy of the model, the overall detection accuracy is shortened, and the overall risk detection accuracy is improved along with the improvement of the system, and the overall detection accuracy is more suitable for the defect detection time is shortened.

It can be understood that by inputting the gradient detection data under the irradiation of each round of light source into the feedback model and adjusting the model parameters, the recognition accuracy of the wafer defect can be effectively improved, particularly under the condition of complex and large illumination condition change, the feedback model has the capability of self-learning and optimization by continuously absorbing the gradient data and adjusting the parameters, can adapt to new process requirements or environment change, can recognize conventional defects through the gradient data and the parameter adjustment, can also process complex and tiny defects, improves the comprehensiveness and accuracy of detection, and can recognize the defects in the production line in real time and efficiently, reduce missing detection and false detection, improve the quality and efficiency of the whole production.

Specifically, in step S5 of the embodiment provided by the present invention, based on the wafer detection feedback model adjusted by the parameters, the detection effect of the existing light source conditions on the wafer surface defect is analyzed, which includes evaluating the indexes such as defect detection accuracy, omission rate, false detection rate and the like under different light source conditions, and determining which light source parameters (such as wavelength, light intensity, angle and the like) have a larger influence on the identification of the wafer defect by modeling the optical characteristics (such as reflectivity, transmissivity, glossiness and the like) of the wafer, which is helpful to find the defect of the current light source configuration, and determine the light source characteristics to be expanded.

More specifically, the requirement analysis of expanding the light source is performed based on the defect type, the wafer material, the surface treatment process and other dimensions, and the light source type and the corresponding parameters thereof which are specifically required to be increased, for example, the light sources with specific wavelengths and different illumination angles or the light source configuration of multi-band combined detection may be required to be increased. Finally, a complete detection scheme capable of covering different light source conditions is designed, the detection bottleneck of the current light source conditions can be comprehensively known through demand analysis, a clear direction is provided for subsequent scheme design, appropriate light source parameters are determined according to different defect types and different characteristics of wafer materials, and an extended light source detection scheme with high adaptability is designed.

More specifically, based on the result of the demand analysis, light sources with different wavelengths, different intensities and different incident angles are selected for combination, the light sources include various types of ultraviolet light, infrared light, white light and the like, which light sources are specifically selected, materials, defect types and production process requirements of a wafer need to be combined, the layout of the light sources is designed according to the size and shape of the wafer, the light sources can be ensured to uniformly irradiate the surface of the wafer, generally, the distribution mode (such as annular shape, radiation type and the like) of the light sources and the change of irradiation angles are considered to capture detailed information of different areas of the surface of the wafer, and corresponding detectors (such as photodiodes, CCD sensors and the like) are designed according to the selected light source configuration so as to capture light signals reflected from the surface of the wafer. The number and the positions of the detectors are matched with the light sources to obtain the optimal detection result, and in the expanding light source scheme, the synchronous or alternate work of multiple light sources may need to be realized, so that each light source can comprehensively detect the defects of the wafer surface in different time periods and different positions, and at the moment, the working frequency and the synchronous mechanism of the light sources need to be coordinated to avoid interference.

More specifically, through designing multiple different types of light sources and layout schemes, the visibility of defects on the surface of the wafer can be improved, especially defects (such as micro cracks and surface scratches) which are not easy to detect under a conventional light source can be more accurately identified, through accurately designing the layout of the light sources, the omnibearing coverage of each area on the surface of the wafer is ensured, and the problem of missed detection caused by uneven coverage of the light sources is reduced.

More specifically, according to the detection scheme of the extended light source, different light sources are started, the wafer is detected round by round or synchronously, in the detection process of each round, the wavelength, the intensity and the incidence angle of the light source are adjusted according to the designed light source parameters, in the detection process of each round, the detector is used for collecting the light signal data reflected from the surface of the wafer, the data generally comprise information such as light intensity, reflected images and transmitted images, the collected data are preprocessed, including denoising, standardization, contrast adjustment and the like, so as to ensure the data quality, the extension data detected in each round are stored in a database, the corresponding light source conditions (such as the light source type, the wavelength and the incidence angle) are marked, the optical detection data (such as the reflection data and the transmission data under different light sources) with multiple dimensions are obtained through the use of the extension light source, the collected extension data can effectively supplement the defect characteristics possibly missed under the conventional light source, and the comprehensive identification capability of the whole detection system is improved.

More specifically, the multi-dimensional detection data obtained from the detection of the extended light source is input into the regulated wafer detection feedback model, at this time, the model receives the extended data including information such as the type of the light source, the wavelength, the incident angle and the like, the model performs defect identification and classification based on the extended detection data, and defect information of each detection area is output, and the results may include defect type, defect position, defect size and the like, and according to the defect identification result fed back by the extended detection data, the model can perform further self optimization and adjustment. For example, if the detection result under a certain light source condition is not ideal, the model can adjust related parameters according to feedback so as to improve the accuracy of the next round of detection, and the model can help the model to better understand the defect characteristics under different light source conditions by substituting the expanded light source data into the model, so that the accuracy and the comprehensiveness of defect identification are improved, and the model can continuously adjust and optimize itself through feedback learning, adapt to various light source configurations and improve the robustness and the stability in different production environments.

It can be understood that by expanding the light source detection scheme, the model can comprehensively collect the defect data of the wafer surface, including the tiny defects which are difficult to detect under different light source conditions, so that the detection precision is remarkably improved. The combination of the light source detection scheme and the model feedback mechanism is expanded, so that the detection system can be flexibly adapted to different types of wafers, different defects and different production environments. Through multiple light source configurations and data feedback, the system can more comprehensively identify various defects, particularly defects which are difficult to find under a conventional light source, such as surface microcracks, transparency changes and the like, in summary, through expanding light source detection requirement analysis, designing an expanded light source detection scheme and combining with a wafer detection feedback model, the accuracy, the comprehensiveness and the adaptability of wafer detection can be greatly improved, the detection efficiency of a production line is optimized, and the phenomena of missing detection and false detection are reduced.

Preferably, the step of irradiating the wafer to be detected with an initial light source in a surrounding state by a preset ring light source and collecting image data of the wafer to be detected under irradiation of the initial light source in the surrounding state to obtain initial detection data of the wafer to be detected includes:

S11, configuring initial brightness parameters and initial orientation parameters of all light source units in a preset annular light source, so that all the light source units in the annular light source are in an initial irradiation state, and all the light source units in the initial irradiation state irradiate the initial light source in a surrounding state on the wafer to be detected;

S12, carrying out parameter configuration of shooting positions and shooting modes on a plurality of preset shooting modules so that each shooting module can acquire image data of the wafer to be detected, and obtaining detection images of the wafer to be detected, which are obtained by the shooting modules in the shooting modes, at the shooting positions, wherein the detection images jointly form initial detection data of the wafer to be detected.

Specifically, according to the characteristics (such as materials and surface reflection characteristics) of the wafer to be detected, the initial brightness parameters of each light source unit in the annular light source are adjusted, the output power of each light source unit can be adjusted, the brightness of the light source can be ensured to irradiate the surface of the wafer, overexposure is avoided, the irradiation angle of each light source unit is adjusted, the light source uniformly irradiates the surface of the wafer from different angles, shadows and dead angles caused by irradiation of the light source with a single angle are avoided, the visibility of defects is improved, each light source unit is configured to be in an initial irradiation state, the light source can uniformly and continuously irradiate the wafer to form surrounding initial light source irradiation conditions, uniform irradiation of each area of the surface of the wafer can be ensured by accurately setting the brightness and the orientation of the light source unit, the missed detection problem caused by the shadows and the nonuniform irradiation is reduced to the greatest extent, the irradiation mode can be flexibly adjusted according to different wafer materials and surface optical characteristics by adjusting the brightness and the angles of the light source, and the defect identification effectiveness is improved.

More specifically, the mounting positions of the camera modules are determined to ensure that the wafer can be comprehensively photographed from a plurality of angles, camera modules at different positions (such as the inside and the outside of the annular light source, different photographing angles and the like) are arranged according to the layout of the annular light source, the surface of the wafer is ensured to be comprehensively covered, the working modes of each camera module are arranged according to detection requirements, the working modes generally comprise resolution, focusing mode, exposure time, image acquisition rate and the like, the different camera modes can help capture defects of different sizes and different characteristics, the diversity and the comprehensiveness of data are improved, the camera modules are ensured to synchronously synchronize or acquire image data on the surface of the wafer at different angles and different modes under the configuration of the different camera modules, at the moment, the working parameters of all camera modules are coordinated in turn, mutual interference is avoided, the image is ensured to be acquired from a plurality of angles through reasonable layout of the camera modules, the surface condition of the wafer is comprehensively reflected, the proper exposure, the resolution and the focusing of each camera module is ensured to acquire image data with high quality under the accurate camera mode parameter setting, and a more reliable basis is provided for the subsequent analysis of the image.

More specifically, according to the set image capturing positions and modes, a plurality of image capturing modules can simultaneously or sequentially capture images of the surface of the wafer, each image capturing module can acquire image data in a corresponding position and a specific mode, the acquired image data can be subjected to necessary preprocessing, such as denoising, image enhancement, brightness equalization and other operations, the image definition and detail distinction are ensured, the detection image data from different image capturing modules are combined to form complete initial detection data, the initial detection data comprises multi-angle and multi-view image information of the surface of the wafer under the irradiation of different angles and different light sources, the state of the surface of the wafer is comprehensively reflected, the parameters of the image capturing modules are accurately configured, the image capturing modules can be ensured to acquire clear images with high resolution, the defect recognition precision is improved, the image data can cover all areas of the surface of the wafer through the cooperation of the plurality of image capturing modules, and abundant information is provided for the subsequent defect analysis.

More specifically, the image data obtained from different positions and under different light source irradiation are summarized to form preliminary detection data, the data is used as input of subsequent defect detection, defect classification and model optimization, preliminary analysis is carried out on the preliminary detection data, possible surface defects and defect areas are identified, the information can provide references for subsequent defect diagnosis and repair processes, the preliminary detection data comprises comprehensive information of the wafer surface through summarizing multi-angle and multi-mode image data, a more valuable basis can be provided for subsequent analysis, and the high quality and the comprehensiveness of the preliminary detection data enable the subsequent defect identification model to more accurately identify and classify the tiny defects of the wafer surface.

It can be understood that through the arrangement of the annular light source, the wafer surface is ensured to be uniformly and omnidirectionally irradiated, the problem of missed detection caused by nonuniform irradiation of the light source is reduced, the cooperation of a plurality of camera modules can comprehensively cover the wafer surface, dead angles are avoided, the accuracy of defect identification can be improved through different modes, the collected multidimensional image data can provide more information for subsequent defect analysis, the system is ensured to be suitable for different types of wafers and defects, and due to the high quality and omnidirectionality of initial detection data, the subsequent defect detection system can utilize the data to perform more accurate defect analysis and judgment, so that the overall detection efficiency and accuracy are improved.

Preferably, the step of performing digital simulation on the wafer to be detected according to the initial detection data of the wafer to be detected to obtain the wafer detection feedback model, and performing gradient scheme analysis on the subsequent light source irradiation requirement of the annular light source based on the wafer detection feedback model to obtain a light source irradiation gradient execution scheme includes:

S21, analyzing the initial detection data of the wafer to be detected to obtain wafer basic information and wafer detection information of the wafer to be detected, wherein the wafer basic information is used for describing the wafer specification and the wafer shape of the wafer to be detected, and the wafer detection information is used for describing the detection condition displayed on the wafer surface of the wafer to be detected;

S22, carrying out digital simulation on the wafer to be detected according to the wafer basic information of the wafer to be detected to obtain a wafer detection feedback model for carrying out digital twin simulation feedback on the wafer to be detected, wherein the wafer detection feedback model comprises a wafer substrate simulation part and a wafer surface simulation part;

s23, substituting the wafer detection information into the wafer detection feedback model to perform model parameter adjustment on a wafer surface simulation part of the wafer detection feedback model according to the wafer detection information, so as to perform digital simulation feedback on the wafer surface detection condition of the wafer to be detected through the wafer surface simulation part after the model parameter adjustment;

S24, analyzing the quality of the wafer at each specific position on a wafer surface simulation part of the wafer detection feedback model to obtain detection quality characteristic distribution of the wafer detection feedback model, and extracting and marking defect characteristics of the wafer detection feedback model according to the detection quality characteristic distribution to obtain a defect marking set on the wafer detection feedback model, wherein the defect marking set is used for digitally feeding back the defects on the wafer to be detected on the wafer detection feedback model;

S25, analyzing light source detection effects on specific positions of the wafer surface simulation part of the wafer detection feedback model based on the initial light source irradiation so as to obtain a light source detection defect type adaptation range corresponding to the detection quality characteristic distribution of the wafer detection feedback model;

S26, performing differential analysis on the light source detection defect type adaptation range according to a preset defect type standard to be detected so as to obtain a plurality of light source detection defect type adaptation ranges to be executed subsequently, and performing working parameter analysis and execution gradient arrangement of light source execution requirements on the annular light source according to the plurality of light source detection defect type adaptation ranges to be executed subsequently so as to obtain a light source irradiation gradient execution scheme of the annular light source corresponding to the plurality of light source detection defect type adaptation ranges to be executed subsequently.

Specifically, the method comprises the steps of establishing a digital model of the wafer by using data including geometric characteristics (such as diameter, thickness, curvature and the like) of the wafer, ensuring that the simulation is consistent with the shape and the size of an actual physical wafer, and relating to the detection condition (such as defect type, defect distribution, surface roughness and the like) displayed on the surface of the wafer, wherein the information is used for describing the current state of the surface of the wafer, providing references for subsequent simulation and feedback adjustment, ensuring that the basic information of the subsequent simulation completely reflects the characteristics and the current state of the actual wafer by accurately analyzing initial data, providing reliable basis for subsequent operation, combining the geometric information of the wafer with the detection condition, and ensuring that the physical characteristics of the wafer and the surface defects of the wafer can be considered in the simulation process.

More specifically, the wafer to be detected is subjected to digital twin modeling through a digital simulation technology (based on wafer basic information), a wafer detection feedback model for detection feedback is generated, the model comprises two main parts, namely a wafer substrate simulation part simulates a basic physical structure of the wafer, including material characteristics, thickness and the like of the wafer, a wafer surface simulation part simulates the characteristics of morphology, defects and the like of the wafer surface according to initial detection data, the physical state and the surface defects of the wafer can be accurately reproduced through the digital twin technology, a precise virtual platform is provided for subsequent light source and defect analysis, the simulated model can feed back the change of the wafer surface in real time, and therefore data support is provided for optimizing a light source irradiation strategy.

More specifically, the actually collected wafer detection information is substituted into a feedback model, particularly a wafer surface simulation part, the surface condition of the wafer is reflected more accurately by adjusting simulation parameters (such as distribution, depth and the like of surface defects), accurate simulation feedback can be made according to the actual wafer surface state (such as defect type, distribution and the like) by adjusting the model parameters in real time, the accuracy of a simulation result is improved, the adjusted model is more close to the detection condition of the actual wafer, and more practical value data is provided for subsequent light source irradiation optimization and defect detection.

More specifically, the quality analysis is performed on each specific position of the wafer through the surface simulation part of the simulation feedback model, the detection condition of each position is judged, the defect characteristics are extracted and marked according to the result of the wafer quality analysis, a defect marking set is formed, the marking set reflects the specific position, type and severity of the surface defect of the wafer, the problem area of the surface of the wafer can be accurately identified through the analysis of simulation data and the defect marking, guidance is provided for the follow-up light source irradiation strategy and defect repair, the defect type can be qualitatively judged, the severity of the defect can be quantitatively analyzed, and the follow-up treatment is facilitated.

More specifically, based on initial light source irradiation conditions, light source effect analysis is performed on a wafer surface simulation part in a feedback model, influences of different light source irradiation on wafer surface defects are determined, and by combining light source detection effects, which types of defects can be detected under different light source irradiation conditions and which defects are not easily found by the existing light source conditions are analyzed, and through adaptive analysis of light sources and defect types, optimal light source configuration can be selected for each defect type, so that detection efficiency is maximized, and through analysis of adaptability of the light sources to different defects, optimization of subsequent light source irradiation parameters can be guided, so that defect detection rate is improved.

More specifically, based on preset defect type standards, differential analysis of the application range of the defect types of the light source detection is performed, which light source irradiation conditions are suitable for a certain defect type and which are unsuitable, a gradient scheme of light source irradiation requirements is determined according to analysis results, corresponding light source irradiation gradient execution schemes are arranged according to each defect type application range, the most suitable light source irradiation modes can be adopted for different defect types through execution of the gradient scheme, so that the utilization efficiency of the light source is improved, the optimal light source irradiation can be obtained for each defect type through gradient analysis, the intelligent and the accuracy of detection are improved, unnecessary light source waste is avoided through gradient execution of the light source, meanwhile, the detection efficiency is improved, and the potential influence of overexposure on the surface of a wafer is reduced.

It can be understood that by means of digital simulation and defect type adaptation analysis, accurate light source configuration aiming at different defect types can be achieved, so that detection precision is maximized, manual intervention is reduced, automation and intelligence level of a detection process are improved through digital twin simulation, real-time adjustment and design of a gradient execution scheme, all potential defects on the surface of a wafer can be accurately detected under different light source irradiation conditions through multi-angle and multi-mode simulation feedback and optimization analysis, efficient utilization of a light source is guaranteed based on gradient analysis of light source requirements, unnecessary resource waste is reduced, and overall, the configuration of light source irradiation is more accurate and intelligent through combination of digital simulation and feedback control, and the effect and efficiency of wafer defect detection can be remarkably improved.

Preferably, the step of analyzing the light source detection effect based on the initial light source irradiation on each specific position on the wafer surface simulation part of the wafer detection feedback model to obtain the light source detection defect type adaptation range corresponding to the detection quality feature distribution of the wafer detection feedback model includes:

S251, dividing a wafer surface simulation part of the wafer detection feedback model into a positioning grid coordinate system so as to obtain the positioning grid coordinate system for dividing the wafer surface simulation part into a plurality of positioning grids with positioning coordinates;

S252, carrying out grid disassembly processing on the initial light source irradiation based on the positioning grid coordinate system so as to obtain light source irradiation parameter distribution corresponding to the detection quality characteristic distribution;

s253, carrying out adaptability analysis on detection effects of a plurality of preset wafer defect detection types according to the light source irradiation parameter distribution so as to obtain detection effect adaptability of the light source irradiation parameter distribution corresponding to various wafer defect detection types;

And S254, carrying out combination analysis on the detection effect fitness of the light source irradiation parameter distribution corresponding to various wafer defect detection types to obtain a light source detection defect type adaptation range corresponding to the detection quality characteristic distribution of the wafer detection feedback model.

Specifically, the wafer surface simulation part of the wafer detection feedback model is divided to create a positioning grid coordinate system, and the grid system divides the wafer surface into a plurality of grid areas with positioning coordinates. Each grid region represents a specific position on the surface of the wafer, so that subsequent analysis and evaluation of the light source irradiation effect are facilitated, each position in the model can be accurately related to the actual detection effect through grid division, and the light source detection effect of each grid position can be independently evaluated through a fine division grid coordinate system, so that the positioning and analysis accuracy is improved, and the grid division can help to systematically organize detection data and provide a basis for subsequent light source irradiation parameter distribution analysis.

More specifically, based on the above positioning grid coordinate system, the initial light source irradiation conditions (including light source intensity, irradiation angle, irradiation range, etc.) are subjected to grid disassembly processing, which means that each grid region will have independent light source irradiation parameter allocation, through the disassembly, you can set different light source irradiation conditions for each grid region individually to obtain a complete light source irradiation parameter distribution map, wherein the light source irradiation conditions of each grid position have definite values, the disassembly processing of the light source irradiation ensures that each grid region can obtain accurate light source irradiation according to the requirements of the specific position thereof, which improves the flexibility and accuracy of detection, and different regions may need different intensity light source irradiation, and the disassembly processing can ensure that the light source irradiation of each region is optimized in a targeted manner.

More specifically, the method is characterized in that the preset detection types of the wafer defects are adaptively analyzed according to the light source irradiation parameter distribution of each grid area, specifically, the detection effect of each defect type (such as surface scratches, cracks, bubbles and the like) under different light source irradiation conditions is evaluated, whether the defects can be effectively highlighted or detected by analyzing the light source irradiation through comparing the light source irradiation parameters with the response model of each defect, the detection effect of the defects under specific light source irradiation conditions can be analyzed according to different types of defects, the determination of the light source configuration which is most suitable for a certain defect type is facilitated, and the adaptation degree of different light source irradiation conditions to different defect types can be quantified through adaptively analyzing the detection effect of each defect, so that the data support is provided for the light source configuration.

More specifically, the detection effect fitness for each wafer defect type will be subjected to a combinatorial analysis. The detection effect of each grid area under different light source irradiation is combined, the overall detection quality characteristic distribution can be obtained, and then the proper defect type under each light source irradiation condition is determined, the result of the combined analysis is the light source detection defect type application range, namely, under the specific light source condition, which defect types can be effectively detected, and which defect types can be missed or insufficiently detected, and through the combined analysis, the interaction of a plurality of light source conditions and defect types can be comprehensively considered, an optimal light source configuration scheme is provided, the combined analysis can help to determine the light source irradiation range which is most suitable for various defects, and the overall detection scheme is optimized.

More specifically, the application range of the light source detection defect types corresponding to the detection quality characteristic distribution of the wafer detection feedback model can be determined through the result obtained through the combination analysis, the application range represents the wafer defect types which can be effectively detected under the specific light source irradiation condition, the application range provides a basis for the follow-up light source irradiation execution scheme, more accurate defect detection can be realized, the application range analysis of the light source ensures that each defect type can be effectively detected under the specific light source irradiation, missed detection caused by improper light source configuration is avoided, the efficient utilization of light source irradiation resources can be realized through optimizing the application range of the light source, the detection accuracy is improved, and the energy consumption is reduced.

It can be understood that the whole process from light source irradiation analysis to determination of defect type adaptation range combines the following technical effects that through detailed grid division and light source parameter disassembly, accurate light source adjustment of different wafer areas and defect types is realized, through the adaptability analysis of each defect, the optimal detection effect of different types of defects can be ensured, missing detection and false detection are avoided, the light source adaptation range after combined analysis can provide clear guidance for subsequent detection, the high efficiency of light source configuration and the reliability of detection quality are ensured, and through an accurate light source irradiation gradient execution scheme, the detection efficiency is improved and the energy consumption is reduced.

Preferably, the step of sequentially configuring light source parameters of the annular light source according to the light source irradiation gradient execution scheme, so that the annular light source sequentially applies detection light source irradiation to the wafer to be detected, thereby obtaining gradient detection data of the wafer to be detected under each round of detection light source irradiation includes:

S31, configuring gradient brightness parameters and gradient orientation parameters for each light source unit in the annular light source in sequence according to the light source irradiation gradient execution scheme, so that each light source unit in the annular light source is in a gradient irradiation state, and each light source unit in the gradient irradiation state irradiates the gradient light source in a surrounding state on the wafer to be detected;

S32, carrying out parameter configuration of shooting positions and shooting modes on a plurality of preset shooting modules according to the light source irradiation gradient execution scheme, so that each shooting module carries out image data acquisition on the wafer to be detected, and detecting images of the wafer to be detected, which are obtained by the shooting modules in the shooting modes, at each shooting position, and the detecting images jointly form gradient detecting data of the wafer to be detected.

Specifically, according to the implementation scheme of the light source irradiation gradient, the configuration of gradient brightness parameters and gradient orientation parameters is sequentially carried out on each light source unit in the annular light source, the brightness of each light source unit is changed gradually, the light source irradiation with different brightness can be applied to different positions, so that the brightness of each light source unit irradiating the surface of the wafer is in a gradual change state, the irradiation direction of the light source is adjusted, the different angles of the light rays of each light source unit are ensured, and a gradual change irradiation angle is formed, so that a surrounding type light source irradiation mode is provided, the uniform irradiation of the annular light source on the surface of the wafer can be realized by adjusting the brightness and the gradient of the orientation, meanwhile, the error caused by excessively strong or excessively weak irradiation is avoided, the uniformity of the light source irradiation is ensured, the wafer to be detected can be irradiated at a plurality of angles, and the detection capability of different areas of the surface of the wafer is enhanced, and especially for the wafer with complex geometric shape or special defects.

More specifically, according to the execution scheme of the light source irradiation gradient, the preset image capturing positions and image capturing modes are configured, wherein the positioning of the image capturing modules is adjusted according to the position of the wafer to be detected and the irradiation area of the annular light source, the image capturing modules are used for carrying out image acquisition at a plurality of angles and positions so as to ensure the whole coverage of the surface of the wafer to be detected, the working modes (such as exposure time, focusing, imaging modes and the like) of the image capturing modules are adjusted according to different light source irradiation states so as to optimize the image quality under different light source gradients, each image capturing module is used for carrying out image data acquisition on the wafer to be detected according to the image capturing positions and modes of the image capturing modules, and acquiring the detection images of the wafer under each image capturing position, and all the images jointly form gradient detection data of the wafer to be detected.

More specifically, by configuring a plurality of camera modules and adjusting the positions of the camera modules, image data of a wafer to be detected can be obtained from different visual angles, dead angles possibly caused by a single angle are avoided, comprehensive detection coverage is ensured, parameters such as exposure time, focusing and the like can be adjusted according to illumination conditions by configuring camera modes according to different light source gradients, high quality of images irradiated by different light sources can be ensured, particularly in a low-brightness or high-brightness area, the plurality of camera modules are combined with the change of the illumination gradients of the light sources, details of the surface of the wafer under different illumination conditions, particularly defect information under different brightness and angles can be captured, and detection accuracy and sensitivity are improved.

More specifically, the detected images obtained by each camera module together form gradient detection data of the wafer to be detected, the images comprise wafer surface information under different light source irradiation conditions, the images can be used for analyzing the quality state of the wafer, the images can be further used for defect identification, defect positioning, defect classification and the like after post-processing (such as image splicing, feature extraction and the like), the gradient detection data are formed based on image data under multi-angle and multi-brightness conditions, so that defects of all areas on the wafer surface can be found and recorded, higher-dimensional wafer defect information can be obtained by integrating the image data of a plurality of view angles and different light source irradiation states, more basis is provided for subsequent defect analysis and judgment, and different types of wafer defects such as micro scratches, bubbles, cracks and the like can be efficiently identified through the synthesized gradient detection data, and the detection accuracy is improved especially on the detection of complex surfaces or micro defects.

It can be understood that the brightness and orientation gradient configuration of each light source unit in the annular light source can realize a multi-angle and gradual change irradiation mode, which is critical to the detection of complex wafer surfaces and different types of defects, the position and mode configuration of the camera module ensures that each area of the wafer surfaces can acquire high-quality image data, the omission caused by a single visual angle is avoided, the obtained gradient detection data can comprehensively reflect the wafer surface states through a plurality of camera modules and acquire images from different angles and different irradiation states, the accuracy and reliability of defect identification are improved, and the fine defects on the wafer surfaces, especially in low-contrast areas, can be effectively found through the cooperation of accurate light source irradiation gradients and image acquisition, so that the sensitivity and accuracy of defect detection are optimized.

Preferably, substituting the gradient detection data of the wafer to be detected under each round of detection light sources into the wafer detection feedback model to perform model parameter adjustment on the wafer detection feedback model includes:

s41, substituting gradient detection data of the wafer to be detected under each round of detection light sources into a wafer surface simulation part of the wafer detection feedback model, and enabling the wafer surface simulation part of the wafer detection feedback model to perform data distribution processing on the gradient detection data under each round of detection light sources so that the gradient detection data and specific positions of each round of the wafer surface simulation part are in corresponding relation;

S42, analyzing the gradient detection data according to the light source detection defect type adaptation range corresponding to the gradient detection data to obtain a defect detection result fed back by the gradient detection data, and substituting the defect detection result fed back by the gradient detection data into each specific position of the wafer surface simulation part according to the corresponding relation between the gradient detection data and each specific position of the wafer surface simulation part;

S43, adjusting model parameters of specific positions of the wafer surface simulation part according to the defect detection result fed back by the gradient detection data, and comprehensively analyzing the model parameters adjusted by the defect detection result fed back by the gradient detection data corresponding to each round of specific positions of the wafer surface simulation part to finally determine the model parameters of specific positions of the wafer surface simulation part.

Specifically, gradient detection data obtained by the wafer to be detected under the detection light sources of different rounds are substituted into a wafer surface simulation part in a wafer detection feedback model, the wafer surface simulation part is responsible for simulating the surface characteristics of the wafer according to the model, the aim of the step is to ensure that the detection data can be accurately corresponding to a specific position of the wafer surface simulation part, namely, the specific position of each data point in the gradient detection data is matched with a corresponding position in the simulation part, the accuracy of data distribution is ensured, the relation between each data point and the corresponding position in the simulation model can be ensured by corresponding the gradient detection data to the specific position of the wafer surface simulation part, the condition of data deviation or false correspondence is avoided, the accuracy of the model is improved, the step can ensure that the detection data of each point on the wafer surface is accurately processed, and the step is vital to the follow-up defect analysis and model adjustment.

More specifically, the data analysis is performed according to the application range of the light source detection defect type corresponding to the gradient detection data, this step aims to identify the defect type (such as a crack, a bubble, a scratch, etc.) fed back by each gradient detection data point, obtain the defect detection result of the position according to the analysis result of each data point, and make the defect detection results correspond to the specific positions of the wafer surface simulation part one by one, for example, if a scratch is detected at a certain position, the defect detection result corresponding to the position is fed back to the position of the wafer surface simulation model, by analyzing the gradient detection data and feeding back the defect information, an accurate judgment can be made for the specific defects of each position, which is crucial for the positioning and analysis of the defects, since the gradient detection data are obtained under the irradiation of different light sources, the defect information of the wafer surface under different conditions can be fed back more comprehensively, the missing detection possibly caused by a single light source is avoided, and the analysis and feedback mechanism can capture tiny defects or tiny changes of the wafer surface, and the sensitivity and accuracy of the defect detection are improved.

More specifically, according to the defect detection result of each position, the model parameters of the simulation part of the wafer surface are adjusted, specifically, if a defect is detected at a certain position, the position parameters of the model are adjusted correspondingly, for the defect detection result fed back by each round of gradient detection data, the influence of the defect detection result on different positions of the simulation part of the wafer surface is analyzed comprehensively, the defect feedback of each position is weighted, the optimal model parameters of each position are determined finally, the parameters of the simulation part of the wafer surface are adjusted in real time, so that the model can be adjusted according to the actual defect feedback, the adaptability and the accuracy of the model are improved, the adjustment can ensure that the model can be accurately adjusted when facing defects of different types and different positions, the accuracy loss caused by large-scale adjustment is avoided, the interference of single data can be avoided by carrying out weighted analysis on the defect detection result of each position, and the feedback effect of each data point is considered comprehensively, and the accuracy and the reliability of the adjustment of the model parameters are ensured.

More specifically, through the adjustment and weight analysis, the model parameters of each position of the simulation part of the wafer surface are finally determined, the parameters reflect the state of each point of the wafer surface under the specific light source condition, including defect information and wafer surface characteristics, the finally obtained model has high accuracy through multi-wheel gradient data feedback and model parameter adjustment, the actual detection requirement can be better adapted, the accurate model parameters enable the defect detection system to more efficiently identify and position various defects on the wafer surface, and the reliability and efficiency of the detection process are improved.

It can be understood that various defects on the surface of the wafer can be identified and fed back through accurate gradient detection data analysis, the model is adjusted according to real-time defect feedback, simulation parameters of each detection position are ensured to be more accurate, adaptability of the model is enhanced, the defects of the wafer can be comprehensively and finely identified under the support of multi-round detection data under different light source conditions, robustness and accuracy of a detection system are improved, the adjustment of model parameters is ensured to be finer through comprehensive weight analysis, unnecessary excessive adjustment is avoided, finally, an optimized wafer surface simulation result is obtained, and the overall detection effect is improved.

Preferably, the step of analyzing the detection requirement of the extended light source according to the wafer detection feedback model adjusted by the model parameters to obtain an extended light source detection scheme of the wafer to be detected, and detecting the wafer to be detected according to the extended light source detection scheme to obtain extended detection data and substituting the extended detection data into the wafer detection feedback model includes:

s51, acquiring performance information of a preset expansion light source module, wherein the expansion light source module comprises a laser light source module and a polarized light source module;

S52, performing fixed-point analysis on the potential flaws on the wafer detection feedback model after model parameter adjustment to obtain potential flaw detection points of the wafer detection feedback model;

S53, performing expansion light source detection requirement analysis on the potential flaw detection points according to the performance information of the expansion light source module to obtain expansion light source detection schemes of the potential flaw detection points of the wafer to be detected;

S54, carrying out parameter configuration on the extended light source module according to the extended light source detection scheme so that the extended light source module carries out the extended light source application processing of the appointed form on the wafer to be detected, and carrying out image acquisition on potential flaw detection points of the wafer to be detected, which are subjected to the extended light source application processing of the appointed form, through a preset camera module so as to obtain extended detection data;

S55, substituting the expansion detection data into the specific position corresponding to the potential flaw detection point in the wafer detection feedback model, and enabling the wafer detection feedback model to conduct live analysis on the expansion detection data so as to conduct model parameter adjustment on the wafer detection feedback model, so that digital feedback is conducted on the actual condition of the wafer to be detected.

Specifically, performance information of a preset extended light source module is acquired. The expanding light source module comprises a laser light source module and a polarized light source module, wherein the laser light source module is generally used for focusing and positioning tiny defects on the surface of a wafer, such as cracks, tiny scratches and the like, the polarized light source module is used for detecting reflection characteristics on the surface of the wafer, stress, textures and some hidden defects inside materials can be revealed, different types of light sources can be applied in a targeted mode by acquiring performance information of the light source module, the characteristics of each type of light source can be guaranteed to be most matched with a detection target (the defect on the surface of the wafer), the effectiveness of subsequent detection is improved, and the combination of different light source modules enables the system to detect the wafer in all directions under different optical conditions, so that potential defects can be revealed better.

More specifically, the wafer inspection feedback model with the model parameters adjusted is subjected to fixed-point analysis of potential flaws, and potential flaw detection points are identified, wherein the detection points refer to wafer surface positions where flaws or flaws may exist. The potential flaw areas can be identified through the feedback model, the areas often need further refined detection, possible flaw points can be identified from the whole scanning through fixed point analysis, a clear direction is provided for subsequent deeper detection, indiscriminate detection on the whole wafer surface can be avoided through optimizing selection of detection points, time and resources are saved, and the area with potential flaws is concentrated.

More specifically, according to the performance information of the extended light source module, the extended light source detection requirement analysis is performed on the potential flaw detection points, an extended light source detection scheme of each detection point is formulated, which light source (laser or polarized light) is determined to be used according to the properties of the detection points and the problems (such as cracks, stress and reflection characteristics) to be solved, and how to apply the light source (such as the angle and the intensity of the light source), and according to the analysis, the accurate irradiation and imaging can be performed on different potential flaw positions by selecting a proper light source, so that the use of redundant light sources is avoided, meanwhile, the accuracy of defect identification is improved, and the customized light source use scheme can adopt different detection strategies for different types of defects (such as surface scratches, micro cracks and stress points) to improve the comprehensiveness and the accuracy of detection.

More specifically, according to the detection scheme of the extended light source, parameter configuration is carried out on the extended light source module, so that the extended light source module can carry out light source application processing on a wafer according to set requirements, for example, a laser light source can irradiate the surface of the wafer under a specific angle, a polarized light source can select different polarized angles, image acquisition is carried out on the wafer applied with the extended light source through a preset camera module, after the camera module acquires image data, the image data is used for further analysis, particularly around potential flaw detection points, according to specific requirements of the extended light source, the parameter of the light source module is adjusted so that the optimal light source irradiation mode of each potential flaw can be ensured, and therefore the detection sensitivity and accuracy are improved.

More specifically, the extended detection data (i.e. the image data acquired by the extended light source module and the camera module) are substituted into a wafer detection feedback model, the data correspond to specific positions of potential flaw detection points, the model further analyzes the surface of the wafer in real time according to the data, the model adjusts parameters again according to new data on the basis of analysis, so that the actual condition of the wafer to be detected is digitally fed back, the wafer detection feedback model can perform self optimization after each round of detection through the real-time analysis and the parameter adjustment, the identification capability of wafer defects is continuously improved, and the digital feedback enables detection personnel to acquire the defect information of the wafer in real time, so that decisions and corrections can be made more quickly, and the efficiency and quality of the production process are improved.

It can be understood that by acquiring the performance information of the extended light source module, a light source scheme meeting the requirement of potential flaw detection is formulated, so that different defects can be detected under the optimal condition, the area with the possible defects can be rapidly and accurately positioned by carrying out potential flaw fixed point analysis on the surface of the wafer, an accurate target is provided for subsequent depth detection, the extended light source detection scheme of each potential flaw is customized, efficient detection is ensured to be carried out under the optimal light source condition, a high-quality image is acquired through the camera module after the extended light source is applied, analysis and real-time feedback are carried out through the wafer detection feedback model, model parameters are continuously adjusted, and the detection precision is improved.

Referring to fig. 2, in a second aspect, the present invention provides an adaptive adjustment and control device for brightness of a wafer inspection light source, for implementing the adaptive adjustment and control method for brightness of a wafer inspection light source according to any one of the first aspect, including:

In this embodiment, for specific implementation of each module in the above embodiment of the apparatus, please refer to the description in the above embodiment of the method, and no further description is given here.

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.

Claims

1. A method for adaptively controlling the brightness of a wafer inspection light source, comprising:

Using a pre-set annular light source to illuminate the wafer to be inspected in a surrounding state with an initial light source, and collecting image data of the wafer to be inspected under the illumination of the initial light source in the surrounding state, so as to obtain initial inspection data of the wafer to be inspected;

Digitally simulating the wafer to be inspected according to the initial inspection data of the wafer to be inspected to obtain a wafer inspection feedback model, and performing a gradient scheme analysis of subsequent light source illumination requirements on the annular light source based on the wafer inspection feedback model to obtain a light source illumination gradient execution scheme;

According to the light source illumination gradient execution scheme, the light source parameters of the annular light source are sequentially configured, so that the annular light source sequentially applies the detection light source illumination to the wafer to be inspected, thereby obtaining the gradient detection data of the wafer to be inspected under each round of detection light source illumination;

Substituting the gradient detection data of the wafer to be inspected under each round of detection light source into the wafer detection feedback model to adjust the model parameters of the wafer detection feedback model;

Performing an extended light source detection demand analysis according to the wafer detection feedback model after the model parameters are adjusted to obtain an extended light source detection scheme for the wafer to be detected, and detecting the wafer to be detected according to the extended light source detection scheme to obtain extended detection data and substitute it into the wafer detection feedback model;

The steps of digitally simulating the wafer to be inspected according to the initial inspection data of the wafer to be inspected to obtain the wafer inspection feedback model, and performing a gradient scheme analysis of subsequent light source illumination requirements on the ring light source based on the wafer inspection feedback model to obtain a light source illumination gradient execution scheme include:

Parsing the wafer basic information and wafer detection information of the initial detection data of the wafer to be detected to obtain the wafer basic information and wafer detection information of the wafer to be detected; wherein the wafer basic information is used to describe the wafer specifications and wafer shape of the wafer to be detected, and the wafer detection information is used to describe the detection status displayed on the wafer surface of the wafer to be detected;

Digitally simulating the wafer to be detected according to the wafer basic information of the wafer to be detected to obtain a wafer detection feedback model for performing digital twin simulation feedback on the wafer to be detected; wherein the wafer detection feedback model includes a wafer substrate simulation part and a wafer surface simulation part;

Substituting the wafer detection information into the wafer detection feedback model, so as to adjust the model parameters of the wafer surface simulation part of the wafer detection feedback model according to the wafer detection information, thereby performing digital simulation feedback on the wafer surface detection status of the wafer to be detected through the wafer surface simulation part after the model parameter adjustment;

Analyzing the wafer quality at each specific location on the wafer surface simulation part of the wafer detection feedback model to obtain the detection quality characteristic distribution of the wafer detection feedback model;

Based on the initial light source irradiation, the light source detection effect is analyzed at each specific position on the wafer surface simulation part of the wafer detection feedback model, so as to obtain the light source detection defect type adaptation range corresponding to the detection quality characteristic distribution of the wafer detection feedback model;

Performing a difference analysis on the adaptability range of the light source defect detection type according to the preset defect type standard to be detected, so as to obtain the adaptability range of several types of light source defect detection types to be executed subsequently, and performing working parameter analysis and execution gradient arrangement of the ring light source irradiation requirements according to the adaptability range of several types of light source defect detection types to be executed subsequently, so as to obtain the light source irradiation gradient execution scheme of the ring light source corresponding to the adaptability range of several types of light source defect detection types to be executed subsequently;

The steps of analyzing the light source detection effect of each specific position on the wafer surface simulation part of the wafer detection feedback model based on the initial light source irradiation to obtain the light source detection defect type adaptation range corresponding to the detection quality feature distribution of the wafer detection feedback model include:

Dividing the wafer surface simulation part of the wafer detection feedback model into a positioning grid coordinate system to obtain a positioning grid coordinate system for dividing the wafer surface simulation part into a plurality of positioning grids having positioning coordinates;

Performing grid decomposition processing on the initial light source illumination based on the positioning grid coordinate system to obtain a light source illumination parameter distribution corresponding to the detection quality feature distribution;

Performing adaptability analysis on detection effects of several preset wafer defect detection types according to the light source illumination parameter distribution to obtain the adaptability of detection effects of the light source illumination parameter distribution corresponding to various wafer defect detection types;

The adaptability of the detection effects of the light source illumination parameter distribution corresponding to various wafer defect detection types is combined and analyzed to obtain the adaptability range of the light source detection defect types corresponding to the detection quality feature distribution of the wafer detection feedback model.

2. The method for adaptively controlling the brightness of a wafer inspection light source according to claim 1, characterized in that the steps of irradiating the wafer to be inspected with an initial light source in a surrounding state by a pre-set annular light source, and collecting image data of the wafer to be inspected under the irradiation of the initial light source in the surrounding state to obtain the initial inspection data of the wafer to be inspected include:

The initial brightness parameters and initial orientation parameters of each light source unit in the pre-set annular light source are configured so that each light source unit in the annular light source is in an initial irradiation state, and each light source unit in the initial irradiation state performs initial light source irradiation in a surrounding state on the wafer to be inspected;

The parameters of the camera position and the camera mode are configured for several pre-set camera modules so that each of the camera modules can collect image data of the wafer to be inspected, so as to obtain a detection image of the wafer to be inspected at each of the camera positions obtained by the camera modules in the camera mode, and each of the detection images together constitutes the initial detection data of the wafer to be inspected.

3. The method for adaptively controlling the brightness of a wafer inspection light source according to claim 1 is characterized in that the light source parameters of the annular light source are configured in sequence according to the light source illumination gradient execution scheme, so that the annular light source applies the inspection light source illumination to the wafer to be inspected in sequence, thereby obtaining the gradient detection data of the wafer to be inspected under each round of the inspection light source illumination, the step comprising:

According to the light source illumination gradient execution scheme, the gradient brightness parameters and the gradient orientation parameters of each light source unit in the annular light source are sequentially configured, so that each light source unit in the annular light source is in a gradient illumination state, and each light source unit in the gradient illumination state performs gradient light source illumination in a surrounding state on the wafer to be inspected;

According to the light source illumination gradient execution scheme, the parameters of the camera position and the camera mode are configured for several pre-set camera modules, so that each of the camera modules collects image data of the wafer to be inspected, so as to obtain a detection image of the wafer to be inspected at each of the camera positions obtained by the camera module in the camera mode, and each of the detection images together constitutes the gradient detection data of the wafer to be inspected.

4. The method for adaptively controlling the brightness of a wafer inspection light source according to claim 1, wherein the step of substituting the gradient detection data of the wafer to be inspected under each round of the inspection light source into the wafer inspection feedback model to adjust the model parameters of the wafer inspection feedback model comprises:

Substituting the gradient detection data of the wafer to be inspected under each round of detection light source into the wafer surface simulation part of the wafer detection feedback model, and allowing the wafer surface simulation part of the wafer detection feedback model to perform data allocation processing on the gradient detection data under each round of detection light source, so that the gradient detection data is in a corresponding relationship with the specific positions of each location of the wafer surface simulation part;

The gradient detection data is parsed according to the light source detection defect type adaptation range corresponding to the gradient detection data to obtain the defect detection result fed back by the gradient detection data, and according to the correspondence between the gradient detection data and the specific positions of the wafer surface simulation part, the defect detection result fed back by the gradient detection data is substituted into the specific positions of the wafer surface simulation part;

The model parameters of each specific position of the wafer surface simulation part are adjusted according to the defect detection results fed back by the gradient detection data, and a comprehensive weight analysis is performed on the model parameters adjusted by the defect detection results fed back by the gradient detection data in each round at each specific position of the wafer surface simulation part, so as to finally determine the model parameters of each specific position of the wafer surface simulation part.

5. The method for adaptively controlling the brightness of a wafer inspection light source according to claim 1 is characterized in that the steps of performing an extended light source inspection demand analysis based on the wafer inspection feedback model after model parameter adjustment to obtain an extended light source inspection scheme for the wafer to be inspected, and inspecting the wafer to be inspected according to the extended light source inspection scheme to obtain extended inspection data and substitute it into the wafer inspection feedback model include:

Acquire performance information of a preset extended light source module; wherein the extended light source module includes a laser light source module and a polarized light source module;

Performing a fixed-point analysis of potential defects on the wafer inspection feedback model after the model parameters have been adjusted to obtain a potential defect detection point of the wafer inspection feedback model;

Performing an extended light source detection demand analysis on the potential defect detection points according to the performance information of the extended light source module to obtain an extended light source detection solution for each of the potential defect detection points of the wafer to be detected;

According to the extended light source detection scheme, the extended light source module is configured with parameters so that the extended light source module applies a specified form of extended light source to the wafer to be detected, and an image of a potential defect detection point of the wafer to be detected that is subjected to the specified form of extended light source processing is captured through a preset camera module to obtain extended detection data;

The extended detection data is substituted into the specific position of the potential defect detection point corresponding to the wafer detection feedback model, and the wafer detection feedback model is made to perform real-time analysis on the extended detection data so as to adjust the model parameters of the wafer detection feedback model, thereby providing digital feedback on the actual condition of the wafer to be inspected.

6. An adaptive control device for wafer inspection light source brightness, characterized in that it is used to implement an adaptive control method for wafer inspection light source brightness according to any one of claims 1 to 5, comprising:

An initial detection module is used to irradiate the wafer to be detected with an initial light source in a surrounding state through a pre-set annular light source, and collect image data of the wafer to be detected under the irradiation of the initial light source in the surrounding state, so as to obtain initial detection data of the wafer to be detected;

A scheme analysis module, used for digitally simulating the wafer to be inspected according to the initial inspection data of the wafer to be inspected to obtain the wafer inspection feedback model, and performing a gradient scheme analysis of the subsequent light source illumination requirements on the annular light source based on the wafer inspection feedback model to obtain a light source illumination gradient execution scheme;

A gradient detection module, used to sequentially configure light source parameters of the annular light source according to the light source illumination gradient execution scheme, so that the annular light source sequentially applies detection light source illumination to the wafer to be inspected, thereby obtaining gradient detection data of the wafer to be inspected under each round of detection light source illumination;

A data feedback module, used for substituting the gradient detection data of the wafer to be detected under each round of detection light source into the wafer detection feedback model, so as to adjust the model parameters of the wafer detection feedback model;

The extended detection module is used to perform an extended light source detection demand analysis based on the wafer detection feedback model after model parameter adjustment to obtain an extended light source detection solution for the wafer to be detected, and to detect the wafer to be detected according to the extended light source detection solution to obtain extended detection data and substitute it into the wafer detection feedback model.