CN110426395B - A kind of solar EL battery silicon wafer surface detection method and device - Google Patents

A kind of solar EL battery silicon wafer surface detection method and device Download PDF

Info

Publication number
CN110426395B
CN110426395B CN201910597030.8A CN201910597030A CN110426395B CN 110426395 B CN110426395 B CN 110426395B CN 201910597030 A CN201910597030 A CN 201910597030A CN 110426395 B CN110426395 B CN 110426395B
Authority
CN
China
Prior art keywords
image
pyramid
detected
silicon wafer
layer
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910597030.8A
Other languages
Chinese (zh)
Other versions
CN110426395A (en
Inventor
谢宏威
周聪
陈从桂
谢德芳
李德友
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Weihua Intelligent Technology Co ltd
Original Assignee
Guangzhou University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangzhou University filed Critical Guangzhou University
Priority to CN201910597030.8A priority Critical patent/CN110426395B/en
Publication of CN110426395A publication Critical patent/CN110426395A/en
Application granted granted Critical
Publication of CN110426395B publication Critical patent/CN110426395B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/94—Investigating contamination, e.g. dust
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30148—Semiconductor; IC; Wafer

Landscapes

  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Pathology (AREA)
  • Analytical Chemistry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

本发明公开了一种太阳能EL电池硅片表面检测方法，包括：设置金字塔的层数并且根据所述层数，创建对应获取到的源图像的第一金字塔图像以及对应预设的模板图像的第二金字塔图像；逐层遍历所述第一金字塔图像，同时对应每一层的第一金字塔图像生成感兴趣区域图像；对所述感兴趣区域图像与所述第二金字塔图像的每一层图像进行匹配并输出匹配结果；根据所述匹配结果生成待检测图像；检测所述待检测图像并输出检测结果。本发明能够提高源图像与模板图像的匹配效率，从而提高图像处理效率，进而增强电池硅片的质量检测能力。

Figure 201910597030

The invention discloses a method for detecting the surface of a silicon wafer of a solar EL cell. Two pyramid images; traverse the first pyramid image layer by layer, and generate a region of interest image corresponding to the first pyramid image of each layer at the same time; matching and outputting a matching result; generating an image to be detected according to the matching result; detecting the to-be-detected image and outputting the detection result. The present invention can improve the matching efficiency between the source image and the template image, thereby improving the image processing efficiency, thereby enhancing the quality detection capability of the battery silicon wafer.

Figure 201910597030

Description

Method and device for detecting surface of solar EL battery silicon wafer
Technical Field
The invention relates to the technical field of machine vision, in particular to a method and a device for detecting the surface of a solar EL battery silicon wafer.
Background
The photovoltaic industry is developed rapidly, the market demand is very large, the enterprise competition is intense, and the quality of the solar cell directly determines the efficiency and the service life of the photovoltaic power generation system. The quality problems which limit the light spot conversion efficiency and the service life of the cell silicon wafer comprise the surface contamination defects of the cell silicon wafer.
With the continuous development of machine vision technology, the continuous improvement of image processing technology and the continuous optimization of related algorithms, machine vision inspection has become one of the important means for inspecting the surface of the battery silicon wafer. However, the prior art cannot keep pace with the rapid development of the photovoltaic industry, and in the detection process of the traditional method, a machine needs to process a large amount of graphic calculation, so that the image processing efficiency is still very low, and the quality detection capability of the battery silicon wafer is low.
Disclosure of Invention
Aiming at the technical problems, the invention provides a method and a device for detecting the surface of a solar EL battery silicon wafer, which can improve the matching efficiency of a source image and a template image, so as to accelerate the image processing speed and further enhance the quality detection capability of the battery silicon wafer. The technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for detecting a surface of a solar EL cell silicon wafer, including:
setting the layer number of a pyramid, and creating a first pyramid image corresponding to the acquired source image and a second pyramid image corresponding to a preset template image according to the layer number;
traversing the first pyramid images layer by layer, and generating an interested area image corresponding to the first pyramid image of each layer;
matching the region-of-interest image with each layer of image of the second pyramid image and outputting a matching result;
generating an image to be detected according to the matching result;
and detecting the image to be detected and outputting a detection result.
In a first possible implementation manner of the first aspect of the present invention, the matching the region-of-interest image and each layer of image of the second pyramid image and outputting a matching result includes:
and calculating a similarity value of the region-of-interest image and the second pyramid image.
In a second possible implementation manner of the first aspect of the present invention, the generating an image to be detected according to the matching result includes:
and dividing and eliminating the interference region image from the image to be detected.
In a third possible implementation manner of the first aspect of the present invention, the dividing and excluding the interference region image from the image to be detected includes:
and removing a background region image from the image to be detected according to the area characteristics.
In a fourth possible implementation manner of the first aspect of the present invention, the dividing and excluding the interference region image from the image to be detected includes:
extracting black edge characteristics from the image to be detected;
and carrying out region merging operation and region difference set operation according to the black edge characteristics to eliminate the black edge region image.
In a fifth possible implementation manner of the first aspect of the present invention, the detecting the image to be detected and outputting a detection result specifically includes:
and screening out dirty areas from the image to be detected according to the roundness characteristics, and calculating the dirty area and the number of the dirty areas.
In a sixth possible implementation manner of the first aspect of the present invention, the method for detecting a surface of a solar EL cell silicon wafer further includes:
and judging the processing level of the surface of the battery silicon wafer according to the detection result.
In a second aspect, an embodiment of the present invention provides an apparatus for detecting a silicon wafer surface of a solar EL cell, including:
the layer creating module is used for setting the layer number of the pyramid and creating a first pyramid image corresponding to the obtained source image and a second pyramid image corresponding to the preset template image according to the layer number;
the layer access module is used for traversing the first pyramid images layer by layer and generating an interested area image corresponding to the first pyramid image of each layer;
the image matching module is used for matching the region-of-interest image with each layer of image of the second pyramid image and outputting a matching result;
the image positioning module is used for generating an image to be detected according to the matching result;
and the image detection module is used for detecting the image to be detected and outputting a detection result.
In a first possible implementation manner of the second aspect of the present invention, the image matching module includes:
and the similarity calculation module is used for calculating the similarity value of the region-of-interest image and the second pyramid image.
In a second possible implementation manner of the second aspect of the present invention, the image positioning module includes:
and the image screening module is used for dividing and eliminating the interference region image from the image to be detected.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
according to the method for detecting the surface of the solar EL battery silicon wafer, image data are layered in a pyramid mode, a source image with huge data volume is divided into a plurality of layers which are ordered and easy to process, and therefore layering access to the image can be conducted at the same time, and the speed of traversing the image is improved; meanwhile, the pyramid-shaped layering is beneficial to image matching of layers which are small in data volume and are visited first, and the effect of synchronous image traversal and image matching is achieved, so that the matching efficiency of a source image and a template image is improved, the image processing speed is increased, and the quality detection capability of the battery silicon wafer is enhanced.
Drawings
FIG. 1 is a flow chart of a method for detecting the surface of a silicon wafer of a solar EL cell in an embodiment of the present invention;
FIG. 2 is a structural diagram of a silicon wafer surface detection device for a solar EL cell in an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a method for detecting a surface of a solar EL cell silicon wafer according to an exemplary embodiment of the present invention is shown, including:
s101, setting the layer number of a pyramid, and creating a first pyramid image corresponding to a source image and a second pyramid image corresponding to a preset template image according to the layer number; the template image is a standard solar cell silicon wafer image which is shot by X-rays in a darkroom environment, has no surface dirt and is good in quality.
S102, traversing the first pyramid images layer by layer, and generating an interested area image corresponding to the first pyramid image of each layer;
s103, matching the region-of-interest image with each layer of image of the second pyramid image and outputting a matching result; the matching result comprises information such as row, column coordinates and angles of the matching position, matching scores and the like.
It will be appreciated that the larger the number of layers, the less time is used to find a match. In addition, it is necessary to ensure that the image of the highest layer has enough information (at least four points), if the pyramid layer number is too large, the template is not easy to identify, and if the pyramid layer number is too small, the time for finding the template is increased. Therefore, the computer carries out reasonable layering according to the data size of the image to be processed, and is beneficial to shortening the time required by template identification, thereby improving the image matching efficiency.
Preferably, the pyramid layer number is set to be 2 for most of the cell images for matching.
S104, generating an image to be detected according to the matching result;
preferably, the generating an image to be detected according to the matching result includes:
and obtaining the image to be detected by image processing methods such as translation and rotation affine transformation.
And S105, detecting the image to be detected and outputting a detection result.
According to the method for detecting the surface of the solar EL battery silicon wafer, provided by the embodiment of the invention, image data are layered in a pyramid mode, and a source image with huge data volume is divided into a plurality of layers which are ordered and easy to process, so that the image can be accessed in a layered mode at the same time, and the speed of traversing the image is increased; meanwhile, the pyramid-shaped layering is beneficial to image matching of layers which are small in data volume and are visited first, and the effect of synchronous image traversal and image matching is achieved, so that the matching efficiency of a source image and a template image is improved, the image processing speed is increased, and the quality detection capability of the battery silicon wafer is enhanced.
Preferably, the setting of the number of layers and creating, according to the number of layers, a first pyramid image corresponding to the source image and a second pyramid image corresponding to the preset template image includes:
and performing image processing on the first pyramid image and the second pyramid image by adopting a smoothing filter.
It can be understood that, when creating a pyramid image of each layer, downsampling is involved, and the downsampled image may have jaggies and needs to be processed by a smoothing filter; the gaussian smoothing filter works well but is time consuming, preferably using a small template averaging filter for image processing.
Preferably, the matching the region-of-interest image and each layer image of the second pyramid image and outputting a matching result includes:
and calculating a similarity value of the region-of-interest image and the second pyramid image.
It can be understood that, when calculating the similarity value between the template image and the region-of-interest image (ROI image), a similarity measurement criterion needs to be selected; the similarity metric criteria are SAD (sum of absolute values), SSD (sum of squared differences) and NCC (normalized correlation coefficient), which is the most time consuming but most efficient to calculate and can adapt well to the illumination change, and is preferably chosen.
In this embodiment, in the target detection process, a common method is to set a template, and traverse the whole source image in the form of a sliding window; each sliding generates an ROI image with the size equal to that of the template, and based on a certain measuring mode, a similarity measure value of the template and the current ROI image is calculated. Thus, an image is formed after traversing the whole image, and the position corresponding to the maximum similarity metric value is found, which is the position of the target to be searched. The similarity measurement is utilized to quantify the similarity, and the method has great advantages for quickly identifying the image and positioning the target area image.
Preferably, the generating an image to be detected according to the matching result includes:
and dividing and eliminating the interference region image from the image to be detected.
The pyramid-mode-based image shape template matching method can enable any number of subsequent cells to be detected to be capable of performing translation and rotation to position the same position information as a template image when the position information in the image changes, so that the subsequent segmentation effect is consistent, and the universality and the rapidity of the solar EL cell silicon wafer surface detection method are improved.
Preferably, the dividing and excluding the interference region image from the image to be detected includes:
carrying out gray level binarization processing on the image to be detected;
and removing a background area image from the image to be detected after the gray level binarization processing according to the area characteristics.
In this embodiment, the positioned picture to be processed is subjected to gray level binarization; connecting the pictures with similar gray values into a region, preferably screening the area characteristics because the shape-area characteristics of the silicon wafer are most obvious, and setting an area fixed threshold; corroding the area with the screened characteristics and then performing expansion treatment; the regions after feature screening were subjected to morphological opening operations with rectangles 500 × 500 in length and width.
Preferably, the dividing and excluding the interference region image from the image to be detected further includes:
extracting black edge characteristics from the image to be detected;
and carrying out region merging operation and region difference set operation according to the black edge characteristics to eliminate the black edge region image.
It can be understood that four black edges from top to bottom on the surface of the battery silicon wafer interfere with detection, so that misjudgment is caused, and stains such as surface fingerprints cannot be correctly identified and detected.
In the embodiment, four black edges on the silicon chip of the EL battery are respectively extracted, and then the four areas of the four black edges are subjected to area combination operation one by one to obtain the whole black edge area; performing difference set operation on the silicon chip area and the black edge area of the EL battery to be processed; because affine transformation positioning is carried out based on shape template matching, black edge regions can be accurately segmented for any subsequent number of battery silicon wafers to be detected.
Preferably, the detecting the image to be detected and outputting a detection result includes:
carrying out image enhancement processing on the image to be detected;
carrying out local threshold binarization processing and connected region operation on the image to be detected;
and screening out dirty areas from the image to be detected according to the roundness characteristics, and calculating the dirty area and the number of the dirty areas.
It will be appreciated that the smudging is primarily caused by hands or touch smudging, and is generally characterized by a relatively high degree of roundness, typically relative to line marks, hidden cracks, etc., upon which the surface smudging is screened.
Preferably, the detecting the image to be detected further includes:
and judging the processing level of the surface of the battery silicon wafer according to the detection result.
Specifically, whether the surface of the battery silicon wafer in the source image has the dirty defect is judged according to the dirty area of the surface of the battery silicon wafer and the number of the dirty areas, and when the dirty area or the number of the dirty areas do not meet the preset condition, the battery silicon wafer is judged to have the dirty defect on the surface.
In this embodiment, for the dirty aspect, A, B, C three grade types and fail types are specifically defined:
the method comprises the following steps of firstly, judging the number of dirty areas;
when the number of the dirt is less than or equal to M, entering the next step;
when the number of the dirt is larger than M, judging that the grade D is unqualified;
secondly, judging the dirty area;
when the single area of the dirt is less than or equal to SA, judging the area as A level;
when SA is less than or equal to SB, judging as B level;
when SB is less than or equal to SC, judging as C grade;
when the SC is less than or equal to the dirty single area and less than or equal to SD, judging that the D-grade is unqualified;
the specific SA, SB, SC, SD and M threshold values are defined numerically according to the requirements of quality inspectors of EL battery silicon wafer manufacturers.
In this embodiment, a technician can set a threshold value according to needs on the basis of a basic test standard to personalize the test standard, adapt to the special production standard of an enterprise and the change of the industry, improve the flexibility of the steps of the detection method, and improve the enterprise benefit.
Preferably, the method for acquiring the source image specifically comprises the following steps:
adjusting the vertical height of a camera, placing the camera on a laboratory bench by using a ruler, and adjusting the camera to a visual field with the length and width of the range of 175-177 of the size of a picture to be shot;
adjusting the focal length, placing the silicon wafer to be tested on an experiment table, and adjusting the focal length of a camera until four chamfer grid lines of the silicon wafer can be clearly seen;
eliminating lens distortion, using a square calibration board with the side length of 170mm, and adjusting until all round points on the calibration board can be clearly seen;
and obtaining a source image by extracting the solar cell silicon wafer image to be detected shot by X-ray.
Referring to fig. 2, a device for detecting the surface of a solar EL cell silicon wafer according to an exemplary embodiment of the present invention is shown, including:
the layer creating module 201 is configured to set the number of layers of a pyramid and create a first pyramid image corresponding to the obtained source image and a second pyramid image corresponding to the preset template image according to the number of layers;
the layer access module 202 is configured to traverse the first pyramid images layer by layer, and generate an image of an area of interest corresponding to the first pyramid image of each layer;
the image matching module 203 is configured to match the region-of-interest image with each layer of image of the second pyramid image and output a matching result;
the image positioning module 204 is used for generating an image to be detected according to the matching result;
and the image detection module 205 is configured to detect the image to be detected and output a detection result.
Preferably, the image matching module comprises:
and the similarity calculation module is used for calculating the similarity value of the region-of-interest image and the second pyramid image.
Preferably, the image localization module comprises:
and the image screening module is used for dividing and eliminating the interference region image from the image to be detected.
Preferably, the image screening module includes:
and the background region image screening module is used for removing the background region image from the image to be detected according to the area characteristics.
Preferably, the image screening module includes:
the black edge region image screening module is used for extracting black edge characteristics from the image to be detected; and carrying out region merging operation and region difference set operation according to the black edge characteristics to eliminate the black edge region image.
Preferably, the image detection module includes:
and the roundness detection module is used for screening out dirty areas from the image to be detected according to the roundness characteristics and calculating the dirty area and the number of the dirty areas.
Preferably, the solar EL cell silicon wafer surface detection device further includes:
and the judging module is used for judging the processing level of the surface of the battery silicon wafer according to the detection result.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.

Claims (7)

1.一种太阳能EL电池硅片表面检测方法，其特征在于，包括：1. a solar EL battery silicon wafer surface detection method, is characterized in that, comprises: 设置金字塔的层数并且根据所述层数，创建对应获取到的源图像的第一金字塔图像以及对应预设的模板图像的第二金字塔图像，具体包括：采用平滑滤波器对所述第一金字塔图像和第二金字塔图像进行图像处理；所述层数根据所述源图像的数据量大小设置；所述金字塔的最高层图像至少有四个点；Setting the number of layers of the pyramid and creating a first pyramid image corresponding to the acquired source image and a second pyramid image corresponding to the preset template image according to the number of layers, specifically including: using a smoothing filter for the first pyramid The image and the second pyramid image are subjected to image processing; the number of layers is set according to the data volume of the source image; the top image of the pyramid has at least four points; 逐层遍历所述第一金字塔图像，同时对应每一层的第一金字塔图像生成感兴趣区域图像；Traversing the first pyramid image layer by layer, and generating a region of interest image corresponding to the first pyramid image of each layer; 对所述感兴趣区域图像与所述第二金字塔图像的每一层图像进行匹配并输出匹配结果，具体包括：计算所述感兴趣区域图像与所述第二金字塔图像的相似性值；所述相似性值采用NCC归一化相关系数得到；所述匹配结果包括匹配位置的行、列坐标和角度以及匹配得分；Matching the ROI image and each layer of the second pyramid image and outputting a matching result, specifically including: calculating a similarity value between the ROI image and the second pyramid image; the The similarity value is obtained by using the NCC normalized correlation coefficient; the matching result includes the row, column coordinates and angle of the matching position and the matching score; 根据所述匹配结果生成待检测图像；generating an image to be detected according to the matching result; 检测所述待检测图像并输出检测结果，具体为：根据圆度特征，从所述待检测图像中筛选出脏污区域并计算脏污面积和脏污区域的数量。Detecting the to-be-detected image and outputting the detection result, specifically: according to the circularity feature, screening out the dirty area from the to-be-detected image and calculating the dirty area and the number of the dirty area. 2.如权利要求1所述的太阳能EL电池硅片表面检测方法，其特征在于，所述根据所述匹配结果生成待检测图像，包括：2 . The method for detecting the surface of a silicon wafer of a solar EL cell according to claim 1 , wherein the generating an image to be detected according to the matching result comprises: 2 . 将干扰区域图像从所述待检测图像中分割并排除。The image of the interference area is segmented and excluded from the image to be detected. 3.如权利要求2所述的太阳能EL电池硅片表面检测方法，其特征在于，所述将干扰区域图像从所述待检测图像中分割并排除，包括：3. The method for detecting the surface of a silicon wafer of a solar EL cell according to claim 2, wherein the dividing and excluding the image of the interference area from the to-be-detected image comprises: 根据面积特征从所述待检测图像中剔除背景区域图像。The background area image is removed from the to-be-detected image according to the area feature. 4.如权利要求2或3所述的太阳能EL电池硅片表面检测方法，其特征在于，所述将干扰区域图像从所述待检测图像中分割并排除，包括：4. The method for detecting the surface of a solar EL cell silicon wafer according to claim 2 or 3, wherein the dividing and excluding the image of the interference area from the image to be detected comprises: 从所述待检测图像中提取黑边特征；extracting black edge features from the to-be-detected image; 根据所述黑边特征进行区域合并操作及区域差集操作以排除黑边区域图像。The region merging operation and the region difference operation are performed according to the black border feature to exclude the black border region image. 5.如权利要求1所述的太阳能EL电池硅片表面检测方法，其特征在于，还包括：5. The method for detecting the surface of a silicon wafer of a solar EL cell as claimed in claim 1, further comprising: 根据检测结果判断电池硅片表面的加工水平。According to the test results, the processing level of the surface of the battery silicon wafer is judged. 6.一种太阳能EL电池硅片表面检测装置，其特征在于，包括：6. A solar EL cell silicon wafer surface detection device, characterized in that, comprising: 图层创建模块，用于设置金字塔的层数并且根据所述层数，创建对应获取到的源图像的第一金字塔图像以及对应预设的模板图像的第二金字塔图像；The layer creation module is used to set the number of layers of the pyramid and according to the number of layers, create a first pyramid image corresponding to the acquired source image and a second pyramid image corresponding to the preset template image; 图层访问模块，用于逐层遍历所述第一金字塔图像，同时对应每一层的第一金字塔图像生成感兴趣区域图像；a layer access module, used for traversing the first pyramid image layer by layer, and generating a region of interest image corresponding to the first pyramid image of each layer; 图像匹配模块，用于对所述感兴趣区域图像与所述第二金字塔图像的每一层图像进行匹配并输出匹配结果；所述图像匹配模块，包括：An image matching module for matching the region of interest image and each layer of the second pyramid image and outputting a matching result; the image matching module includes: 相似性计算模块，用于计算所述感兴趣区域图像与所述第二金字塔图像的相似性值；所述相似性值采用NCC归一化相关系数得到；所述匹配结果包括匹配位置的行、列坐标和角度以及匹配得分；A similarity calculation module for calculating the similarity value between the region of interest image and the second pyramid image; the similarity value is obtained by using the NCC normalized correlation coefficient; the matching result includes the row of the matching position, Column coordinates and angles and match scores; 图像定位模块，用于根据所述匹配结果生成待检测图像；an image positioning module, configured to generate an image to be detected according to the matching result; 图像检测模块，用于检测所述待检测图像并输出检测结果。The image detection module is used to detect the to-be-detected image and output the detection result. 7.如权利要求6所述的太阳能EL电池硅片表面检测装置，其特征在于，所述图像定位模块，包括：7. The solar EL cell silicon wafer surface detection device according to claim 6, wherein the image positioning module comprises: 图像筛选模块，用于将干扰区域图像从所述待检测图像中分割并排除。The image screening module is used for segmenting and excluding the interference area image from the to-be-detected image.
CN201910597030.8A 2019-07-02 2019-07-02 A kind of solar EL battery silicon wafer surface detection method and device Active CN110426395B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910597030.8A CN110426395B (en) 2019-07-02 2019-07-02 A kind of solar EL battery silicon wafer surface detection method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910597030.8A CN110426395B (en) 2019-07-02 2019-07-02 A kind of solar EL battery silicon wafer surface detection method and device

Publications (2)

Publication Number Publication Date
CN110426395A CN110426395A (en) 2019-11-08
CN110426395B true CN110426395B (en) 2022-02-11

Family

ID=68410250

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910597030.8A Active CN110426395B (en) 2019-07-02 2019-07-02 A kind of solar EL battery silicon wafer surface detection method and device

Country Status (1)

Country Link
CN (1) CN110426395B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112837302B (en) * 2021-02-09 2024-02-13 广东拓斯达科技股份有限公司 Method and device for monitoring state of die, industrial personal computer, storage medium and system
CN113359049B (en) * 2021-04-29 2022-07-29 张火锐 Method for detecting loss of solid-state lithium battery
CN116091613A (en) * 2023-02-06 2023-05-09 盛视科技股份有限公司 Camera definition detection method
CN116433672B (en) * 2023-06-15 2023-08-25 山东九思新材料科技有限责任公司 Silicon wafer surface quality detection method based on image processing

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004205218A (en) * 2002-12-20 2004-07-22 Nec Tohoku Ltd Device and method of inspecting appearance of inside diameter part
JP2005188991A (en) * 2003-12-24 2005-07-14 Nagoya Electric Works Co Ltd Abnormality detection method for cone-shaped or frustum-shaped bump, abnormality detection device for cone-shaped or frustum-shaped bump
WO2011144374A1 (en) * 2010-05-19 2011-11-24 Robert Bosch Gmbh Method and device for characterising pyramidal surface structures on a substrate
CN107578409A (en) * 2017-09-16 2018-01-12 河北工业大学 A method for detecting the appearance of broken grid defects in solar cells
CN107749057A (en) * 2017-09-16 2018-03-02 河北工业大学 A kind of method of solar battery sheet outward appearance spillage defects detection
CN107768269A (en) * 2017-10-30 2018-03-06 河北工业大学 A kind of method of the dirty defects detection of polysilicon solar battery slice outward appearance
CN107831173A (en) * 2017-10-17 2018-03-23 哈尔滨工业大学（威海） Photovoltaic component defect detection method and system
CN109215026A (en) * 2018-09-29 2019-01-15 广东工业大学 A kind of accurate LED defect inspection method of high speed based on machine vision
CN109711457A (en) * 2018-12-20 2019-05-03 江南大学 A Fast Image Matching Method Based on Improved HU Invariant Moment and Its Application

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6422250B2 (en) * 2014-07-08 2018-11-14 キヤノン株式会社 Image processing method, image processing apparatus, program, and recording medium
CN109801279B (en) * 2019-01-21 2021-02-02 京东方科技集团股份有限公司 Object detection method and device in image, electronic device, storage medium

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004205218A (en) * 2002-12-20 2004-07-22 Nec Tohoku Ltd Device and method of inspecting appearance of inside diameter part
JP2005188991A (en) * 2003-12-24 2005-07-14 Nagoya Electric Works Co Ltd Abnormality detection method for cone-shaped or frustum-shaped bump, abnormality detection device for cone-shaped or frustum-shaped bump
WO2011144374A1 (en) * 2010-05-19 2011-11-24 Robert Bosch Gmbh Method and device for characterising pyramidal surface structures on a substrate
CN107578409A (en) * 2017-09-16 2018-01-12 河北工业大学 A method for detecting the appearance of broken grid defects in solar cells
CN107749057A (en) * 2017-09-16 2018-03-02 河北工业大学 A kind of method of solar battery sheet outward appearance spillage defects detection
CN107831173A (en) * 2017-10-17 2018-03-23 哈尔滨工业大学（威海） Photovoltaic component defect detection method and system
CN107768269A (en) * 2017-10-30 2018-03-06 河北工业大学 A kind of method of the dirty defects detection of polysilicon solar battery slice outward appearance
CN109215026A (en) * 2018-09-29 2019-01-15 广东工业大学 A kind of accurate LED defect inspection method of high speed based on machine vision
CN109711457A (en) * 2018-12-20 2019-05-03 江南大学 A Fast Image Matching Method Based on Improved HU Invariant Moment and Its Application

Also Published As

Publication number Publication date
CN110426395A (en) 2019-11-08

Similar Documents

Publication Publication Date Title
US9171364B2 (en) Wafer inspection using free-form care areas
CN107543828B (en) Workpiece surface defect detection method and system
CN113393426B (en) Steel rolling plate surface defect detection method
CN110426395B (en) A kind of solar EL battery silicon wafer surface detection method and device
CN112037203A (en) Method and system for detection of side surface defects based on registration of complex workpiece outer contour
CN113781415B (en) Defect detection method, device, equipment and medium for X-ray image
WO2007062563A1 (en) On-line automatic inspection method for detecting surface flaws of steel during the pretreatment of the ship steel
JP2018522293A (en) Visual identification method of multi-type BGA chip based on linear clustering
CN110443278B (en) Method, device and equipment for detecting thickness abnormality of grid line of solar cell
CN114331995B (en) A real-time positioning method based on multi-template matching based on improved 2D-ICP
CN115020267A (en) Semiconductor surface defect detection method
TW201300769A (en) Profile-based defect detection using a test device
CN116012292A (en) Wafer appearance defect detection method based on machine vision
CN115393290B (en) Edge defect detection method, device and equipment
CN113781458A (en) A recognition method based on artificial intelligence
CN117495846B (en) Image detection method, device, electronic device and storage medium
CN119723200A (en) Defect type detection method and device, electronic device and storage medium
CN114120318B (en) A method for accurate extraction of target points in dark field images based on integrated decision tree
TWI777307B (en) Method, computer program, and computer readable medium of using electroluminescence images to identify defect of solar cell based on deep learning technology
KR101782366B1 (en) Vision inspection method based on learning data using normalization of sample scale
CN118858293A (en) Graphite Sheet Component Appearance Defect Detection Method Based on Photometric Stereo Algorithm
CN117805142A (en) Defect detection method, defect detection apparatus, and computer-readable storage medium
CN117433966A (en) A non-contact measurement method and system for particle size of grinding particles
CN117690846A (en) Solar silicon wafer visual inspection methods, devices, equipment and storage media
CN112581424B (en) Classification extraction method for surface and subsurface defects of optical element

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20250121

Address after: Room 102, No. 7 Tengfei 1st Street, Huangpu District, Guangzhou City, Guangdong Province 510000

Patentee after: Guangdong Weihua Intelligent Technology Co.,Ltd.

Country or region after: China

Address before: No. 230, Waihuan West Road, Guangzhou University Town, Panyu, Guangzhou City, Guangdong Province, 510006

Patentee before: Guangzhou University

Country or region before: China

TR01 Transfer of patent right