CN114463302B

CN114463302B - Processing method and processing device for assembly

Info

Publication number: CN114463302B
Application number: CN202210106800.6A
Authority: CN
Inventors: 唐庆圆
Original assignee: Zhijian Technology Co ltd
Current assignee: Tang Qingyuan
Priority date: 2022-01-28
Filing date: 2022-01-28
Publication date: 2025-07-22
Anticipated expiration: 2042-01-28
Also published as: CN114463302A

Abstract

The embodiment of the present disclosure provides a processing method and a processing device for an assembly. In the processing method, a group of images captured for each assembly during the process of assembling multiple assemblies is obtained. The group of images includes images captured for the assembly at the end of each process in multiple specified processes. Qualified assemblies and unqualified assemblies are determined from the multiple assemblies. Then, the corresponding images of the unqualified assembly and the qualified assembly are compared to determine the features that cause the unqualified assembly to be unqualified.

Description

Processing method and processing device for assembly

Technical Field

Embodiments of the present disclosure relate to the field of computer technology, and in particular, to a processing method and a processing apparatus for an assembly.

Background

The assembly is typically composed of multiple components. In some cases, the assembly may include hundreds or thousands of parts. The factories that perform assembly operations on an assembly typically do not produce components, but rather purchase components from suppliers of the individual components (thus, components may also be referred to herein as stock or raw materials). Corresponding processes may then be performed by a plurality of apparatuses in a pipelined manner to assemble the components into an assembly.

In some cases, the finished product that can be used by an end user (e.g., a cell phone) can be referred to as an assembly. In other cases, the assembly may not be a finished product, but rather a module in a finished product. For example, a display module, a communication module, etc. in a cell phone may also be referred to as an assembly.

In production management, in order to avoid supply shortages of incoming material or in view of incoming material costs, it is often possible that the same component is provided by at least two suppliers. The tolerances of the components from different suppliers may be different. For example, the tooling of a more costly component may be finer and thus the tolerances of the component may be smaller. Component tolerances can be understood as production errors of the component, e.g. dimensional errors, color errors, brightness errors, etc. Furthermore, in order to avoid line malfunctions or for equipment cost considerations, it is often possible for the same piece of equipment to be assembled on at least two lines. The equipment performing the same process on these pipelines may come from different equipment vendors, and thus the tolerances of the same process performed by different equipment may also be different. For example, more costly equipment may operate more accurately and thus the tolerances of the process it performs may be smaller. The process tolerances are understood to be errors in the accuracy of the assembly, such as errors in the angle of connection between the two components, etc.

Disclosure of Invention

Embodiments described herein provide a processing method, processing apparatus, and computer-readable storage medium storing a computer program for an assembly.

According to a first aspect of the present disclosure, a method of handling an assembly is provided. In the processing method, a set of images acquired for each assembly during the assembly of a plurality of assemblies is acquired. Wherein the set of images includes images acquired for the assembly at the end of each of the plurality of designated processes. A pass assembly and a fail assembly of the plurality of assemblies are determined. The corresponding images of the failed assembly and the pass assembly are then compared to determine the features that caused the failed assembly to fail.

In some embodiments of the present disclosure, the processing method further includes determining a component of the assembly associated with the feature, determining a first feature of the feature associated with a dimensional error of the component, and determining a tolerance range of the component from the image of the failed assembly and the pass assembly including the first feature.

In some embodiments of the present disclosure, the processing method further includes determining a set of associated features of the features, determining a plurality of components of the assembly associated with the set of associated features, determining a first feature of the set of associated features associated with dimensional errors of the plurality of components, and determining a tolerance range for each of the plurality of components based on the image of the failed assembly and the qualifying assembly including the first feature and the cost of the plurality of components.

In some embodiments of the present disclosure, the processing method further includes determining a component of the assembly associated with the feature, determining a second feature of the feature associated with an assembly error of the component, determining a target process of the plurality of specified processes associated with the second feature, and determining a tolerance range of the target process from the image of the failed assembly and the qualified assembly including the second feature.

In some embodiments of the present disclosure, the processing method further includes determining a set of associated features of the features, determining a plurality of components of the assembly associated with the set of associated features, determining a second feature of the set of associated features associated with assembly errors of the plurality of components, determining a plurality of target processes of the plurality of designated processes associated with the second feature, and determining a tolerance range for each of the plurality of target processes based on an image of the failed assembly and the qualified assembly including the second feature and a cost of equipment performing the plurality of target processes.

In some embodiments of the present disclosure, the processing method further includes determining a set of associated features of the features, determining a plurality of components of the assembly associated with the set of associated features, determining a first feature of the set of associated features associated with dimensional errors of the plurality of components, determining a second feature of the set of associated features associated with assembly errors of the plurality of components, determining a plurality of target processes of the plurality of designated processes associated with the second feature, determining a tolerance range for each of the plurality of components and a tolerance range for each of the plurality of target processes based on the image of the failed assembly and the image of the qualified assembly including the first feature and the second feature, the cost of the plurality of components, and the cost of the apparatus performing the plurality of target processes.

In some embodiments of the present disclosure, the processing method further includes determining a component of the assembly associated with the feature, determining a third feature of the feature associated with a size of the component, and determining the size of the component from the images of the failed assembly and the pass assembly that include the third feature.

In some embodiments of the present disclosure, comparing corresponding images of a failed assembly to a qualified assembly to determine features that result in failed assemblies is performed by a plurality of deep learning models, a first one of the plurality of deep learning models configured to compare differences between a set of images acquired for the same assembly to determine regions of interest in the images acquired at the end of each process, and a plurality of second one of the plurality of deep learning models configured to compare the same region of interest in the corresponding images of the plurality of assemblies for a plurality of designated processes, respectively, to determine features that result in failed assemblies.

In some embodiments of the present disclosure, the plurality of second deep learning models are trained via respective images for the plurality of test assemblies that have been labeled as unacceptable features and acceptable features, respectively.

According to a second aspect of the present disclosure, a processing device for an assembly is provided. The processing means comprises at least one processor and at least one memory storing a computer program. The computer program, when executed by at least one processor, causes the processing means to carry out the steps of the method according to the first aspect of the present disclosure.

According to a third aspect of the present disclosure, there is provided a computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the method according to the first aspect of the present disclosure.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following brief description of the drawings of the embodiments will be given, it being understood that the drawings described below relate only to some embodiments of the present disclosure, not to limitations of the present disclosure, in which:

FIG. 1 is an exemplary flow chart of a method of processing for an assembly according to an embodiment of the disclosure;

FIG. 2 is a schematic view of an assembly acquired at the end of each of a plurality of designated processes;

FIG. 3 is an exemplary flowchart of steps further included in a processing method according to an embodiment of the present disclosure;

FIG. 4 is another exemplary flowchart of steps further included in a processing method according to an embodiment of the present disclosure;

FIG. 5 is yet another exemplary flowchart of steps further included in a processing method according to an embodiment of the present disclosure;

FIG. 6 is yet another exemplary flowchart of steps further included in a processing method according to an embodiment of the present disclosure;

FIG. 7 is yet another exemplary flowchart of steps further included in a processing method according to an embodiment of the present disclosure, and

FIG. 8 is a schematic block diagram of a processing device for an assembly according to an embodiment of the invention.

Elements in the figures are illustrated schematically and not drawn to scale.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings. It will be apparent that the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments, which can be made by those skilled in the art based on the described embodiments of the present disclosure without the need for creative efforts, are also within the scope of the protection of the present disclosure.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the presently disclosed subject matter belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the specification and relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein. Terms such as "first" and "second" are used merely to distinguish one component (or portion of a component) from another component (or another portion of a component).

As mentioned above, in actual production, the same components used to assemble the same type of assembly may come from different suppliers and therefore the components may have different tolerances. Excessive tolerances in the components can result in failure of the assembly. The equipment used to perform the same process (equipment that adds the same components in the same type of assembly to the assembly) may also come from different equipment vendors and thus the process may have different tolerances. Excessive process tolerances can also lead to failure of the assembly. Further, since there are tolerances in both the raw materials and the process itself, there is a possibility that a combined tolerance due to the tolerances of the raw materials and the process occurs. Too large assembly tolerances can also lead to failure of the assembly. The assembly tolerances of the intermediate products obtained after the individual assembly steps (processes) cannot be predicted before assembly, so that the quality of the final assembly becomes difficult to guarantee. Particularly, when a new product is developed, the influence of new institution design, new materials and new technology on the product qualification rate can be comprehensively amplified, and the problem that the product qualification rate does not reach the standard can be possibly encountered. How to quickly find the cause of the problem and to formulate corresponding control measures and specifications are necessary processes for ensuring the later mass production. This process requires significant effort from experienced engineers into the development of new products.

Some factories may collect some data of part of the process during assembly and store it in, for example, an Excel table. Such as machine-collectable operating parameters and production process parameters. Analysis of large amounts of data by means of Excel tables is very inefficient, generally takes a long time and often does not allow for a quick and systematic association of individual assembly processes. Furthermore, these data are not typically collected for each raw material in each process, and thus the combined effects of each process and raw material may not be able to be quickly traced back to determine the cause of failed assemblies.

Further, in the actual production of the product (assembly of the components into an assembly), not only the quality of the product but also the production cost are considered. To obtain an optimal combination of product quality and cost, a large number of experienced engineers are typically required to conduct the experiments and analysis of the system. This typically requires a lot of manpower and material resources in a real world scenario. And in some scenarios even due to resource limitations, sufficient experimentation and analysis may not be possible. If nearby products are marketed, the time left for engineers to conduct experiments and analysis is more intense.

Embodiments of the present disclosure provide a processing method for an assembly. Fig. 1 illustrates an exemplary flowchart of a method 100 of processing for an assembly according to an embodiment of the present disclosure.

At block S102 of fig. 1, a set of images acquired for each assembly during assembly of the plurality of assemblies is acquired. Wherein the set of images includes images acquired for the assembly at the end of each of the plurality of designated processes. FIG. 2 shows a schematic view of an assembly acquired at the end of each of a plurality of designated processes. The plurality of designated processes may be all processes in the assembly process, may be continuous processes, or may be processes that have a critical impact on the yield of the assembly.

The image acquired of the console before the assembly operation is initiated is shown at block 210 of fig. 2. In some embodiments of the present disclosure, an image of the console prior to the beginning of the assembly operation may be acquired for comparison with an image acquired at the end of the first designated procedure. Although any components are not shown in block 210, in actual production, the image acquired for the console may reflect image information such as the reflection of light or background color on the console. In other embodiments of the present disclosure, a substrate for carrying other components may be included on the console. In the example of block 210, the substrate may be white and therefore not apparent in the figure. In still other embodiments of the present disclosure, the image shown at block 210 may not be acquired.

The image acquired at the end of the specified process I is shown at block 220. As shown in block 220, component A is acquired in a designated process I, which is positioned in the upper left corner of the console. The image acquired at the end of the specified process II is shown at block 230. As shown in block 230, component A' is acquired in designated process II, which is positioned in the upper right hand corner of the console.

The image acquired at the end of the specified process III is shown at block 240. As shown in block 240, component B is acquired in a designated process III, which is centered on the stage. The image acquired at the end of the specified process IV is shown at block 250. As indicated at block 250, component C is acquired in a specified process IV, which is placed directly under and connected to component B. Similarly, blocks 260 through 290 illustrate images acquired at the end of the designated process V through the designated process VIII, respectively.

As described above, a set of images as shown in fig. 2 is acquired for each assembly. The set of images may be stored in the storage device in association with the number of the assembly.

Returning to fig. 1, at block S104, pass assemblies and fail assemblies of the plurality of assemblies are determined. The manner in which the assembly is determined to be acceptable may be by conventional inspection. A normally functioning assembly is determined to be a pass assembly and a dysfunctioning assembly is determined to be a fail assembly. The result of whether the assembly is acceptable may be stored in the storage device in association with the number of the assembly.

At block S106, the corresponding images of the failed assembly and the pass assembly are compared to determine features that lead to failure of the failed assembly. In some embodiments of the present disclosure, comparing corresponding images of failed assemblies to qualified assemblies may be performed by a plurality of deep learning models to determine features that result in failed assemblies being failed.

A first one of the plurality of deep-learning models may be configured to compare differences between a set of images acquired for the same assembly to determine a region of interest in the images acquired at the end of each process. In the example of fig. 2, the first deep learning model may, for example, determine a distinction of images acquired for any two consecutive processes to determine a region of interest in images acquired at the end of a subsequent process in the two consecutive processes. For example, the first deep learning model may compare the images at block 220 and block 230, determine that the image at block 230 is more component a' than the image at block 220 at the upper right corner. Thus, it may be determined that the region of the image at block 230 where component A' is located is a region of interest. Similarly, the first deep learning model may compare the images at block 240 and block 250, determining that the image at block 250 is one more component C below than the image at block 240. Thus, the region of the image at block 250 where component C is located may be determined to be a region of interest.

A plurality of second deep learning models of the plurality of deep learning models may be configured to compare the same region of interest in corresponding images of the plurality of assemblies for a plurality of designated processes, respectively, to determine features that result in failed assemblies. In the example of fig. 2, there may be 8 second deep learning models for comparing the images of each assembly at blocks 220 through 290, respectively. For example, one of the second deep learning models may compare the region of interest of the image of each assembly at block 230 at the upper right corner to determine whether the upper right corner component a' includes features that result in assembly failure. Another second deep learning model may compare the underlying region of interest of the image of each assembly at block 250 to determine whether the underlying component C includes features that lead to assembly failure.

By determining the region of interest in the image and comparing only the features in the region of interest, the computation time and resources of the deep learning model can be saved, and the efficiency can be improved.

The deep learning model may be, for example, a Mask R-CNN model, a B-MR-CNN model, pointRend model, MASK TRANSFINER model, BPR model, REFINEMASK model, BCNet model, or the like.

In some embodiments of the present disclosure, the plurality of second deep learning models may be trained via respective images for the plurality of test assemblies that have been labeled as unacceptable features and acceptable features, respectively. For example, images as shown in blocks 210 through 290 of FIG. 2 may be acquired for multiple test assemblies in advance. The test assembly may be identical to an actual production assembly, such as a pre-test production assembly. It may be determined whether each test assembly is acceptable. Then, whether each feature of each component in each test assembly is a pass feature or a fail feature may be respectively noted for the images shown in blocks 220 through 290. The same second deep learning model may then be trained using the annotated images corresponding to the same process. For example, a second deep learning model may be trained with labeled images as shown in block 220 acquired for each test assembly. Another second deep learning model may be trained with the annotated image acquired for each test assembly as indicated in block 230. And so on.

The trained deep learning model can identify differences between a pass assembly and a fail assembly and determine features that lead to fail assemblies failing.

In some embodiments of the present disclosure, after determining the features that result in reject assemblies, the scheme of manufacturing assemblies may be adjusted based on the features to improve product yields and/or reduce costs. Fig. 3-7 illustrate exemplary flowcharts of steps further included in a processing method according to embodiments of the present disclosure. The steps shown in fig. 3 to 7 are referred to below to describe how to increase the product yield and/or reduce the cost. The steps shown in fig. 3 to 7 may be performed after block S106 in fig. 1.

Hereinafter, it may be assumed that the cases that result in some assembly failure in the example of fig. 2 include:

(1) The top bulge of the component B is too wide to prevent the wireless signal from passing through;

(2) The mismatch in dimensions of part B and part C results in poor contact;

(3) The placing angle of the component A is inaccurate;

(4) Component E is not aligned with component E';

(5) The distance between the component D and the component C is too short (the length of the component D may be too long, or the placement positions of the components C and D may be inaccurate).

In the embodiment of fig. 3, at block S302, components of the assembly associated with features that result in failed assemblies are determined. Under the above assumption, by comparing corresponding images of a pass assembly and a fail assembly, it can be determined that the components of the assembly associated with the feature that caused the fail assembly to fail are component a, component B, component C, component D, component E, and component E'.

At block S304, a first one of the features associated with a dimensional error of the component is determined. In the above examples, the components associated with the features that caused the reject assembly to reject are component a, component B, component C, component D, component E, and component E'. The first of the features that are associated with dimensional errors of the components that result in failure of the failed assembly may include the width of the top protrusion of component B, the width of the bottom protrusion of component B, the width of the groove of component C, and the size of component D. In the above example, it is assumed that the standard value of the width of the top bump of the component B of the reject assembly is 10mm, whereas it is actually 10.5mm, resulting in weakening of the wireless signal through both sides thereof. It can be determined that dimensional errors in the width of the top projection of component B are responsible for failure of the assembly. Assuming that the standard value of the width of the bottom projection of part B of the reject assembly is 10mm, it is actually 10.5mm, resulting in its inability to be inserted into a groove of standard width 10.1mm of part C. It can be determined that dimensional errors in the width of the bottom protrusions of component B are responsible for failure of the assembly. Assuming that the standard value of the width of the groove of part C of the reject assembly is 10.1mm, it is actually 9.5mm, resulting in it not being able to accommodate the projection of part B of standard width 10mm. It can be determined that dimensional errors in the width of the groove of the component C are responsible for the failure of the assembly. The standard value for the distance of part D from part C for a reject assembly is 2.5mm, while it is actually 2.0mm. This is because the length of the component D exceeds the length standard value. It can be determined that dimensional errors in the length of the component D are responsible for the failure of the assembly.

At block S306, a tolerance range of the component is determined from the images of the failed assembly and the qualified assembly including the first feature. In the above example, the widths of the top protrusions of part B of the pass and fail assemblies may be compared by the image at block 240. For example, the width of the top protrusions of part B in a pass assembly is between 9.7mm and 10.3mm, while the width of the top protrusions of part B in a fail assembly is at least 10.4mm. Then the tolerance range for the width of the top bump of component B can be determined to be + -0.3 mm. In other words, when the width of the top protrusion of the component B is less than 10.3mm, it can be made such that it does not affect the performance of the assembly. Therefore, when purchasing component B, for example, component B should be selected to have a tolerance range of ±0.3mm.

In a similar manner to the determination of the tolerance ranges for the widths of the top protrusions of component B, the tolerance ranges for the widths of the bottom protrusions of component B, the tolerance ranges for the widths of the grooves of component C, and the tolerance ranges for the lengths of component D may be determined.

In the above example, the mismatch in the dimensions of the component B and the component C results in poor contact. In some embodiments of the present disclosure, the dimensions of component B and component C may be considered in combination to determine the tolerance ranges for both. Fig. 4 shows a process flow for this case.

At block S402, a set of associated features among the features that result in the failed assembly being failed is determined. In the above example, it may be determined that the dimensional match of part B to part C is a set of associated features (first set of associated features). The positional match of component E and component E' is a set of associated features (second set of associated features). The relative positions of component C and component D and the length of component D are a set of associated features (third set of associated features).

At block S404, a plurality of components in the assembly associated with the set of associated features is determined. In the hypothetical example above, it may be determined that the plurality of components associated with the first set of associated features are component B and component C, the plurality of components associated with the second set of associated features are component E and component E', and the plurality of components associated with the third set of associated features are component C and component D.

At block S406, a first feature of the set of associated features associated with dimensional errors of the plurality of components is determined. In the above hypothetical example, the first feature associated with the dimensional error of the component among the first set of associated features is the width of the bottom protrusion of component B and the width of the groove of component C. The first feature associated with the dimensional error of the component is not referred to in the second set of associated features. The first feature of the third set of associated features that is associated with the dimensional error of the component is the length of the component D.

At block S408, a tolerance range for each of the plurality of components is determined from the images of the failed assembly and the qualified assembly including the first feature and the cost of the plurality of components. In the hypothetical example above, the tolerance ranges for part B and part C can be determined from the images of the pass and fail assemblies at block 250 for the first set of associated features. In an example where the standard value of the width of the bottom protrusion of the component B is 10mm and the standard value of the width of the groove of the component C is 10.1mm, if the tolerance range of the width of the bottom protrusion of the component B is ±0.05mm, the tolerance range of the width of the groove of the component C may be ±0.05mm. If the tolerance range of the width of the bottom protrusion of the component B is + -0.03 mm, the tolerance range of the width of the groove of the component C may be + -0.07 mm. It is assumed that the price of the component B having the tolerance range of ±0.05mm is 10 yuan cheaper than the component B having the tolerance range of ±0.03mm, and the price of the component C having the tolerance range of ±0.05mm is 2 yuan cheaper than the component C having the tolerance range of ±0.07mm. Then the tolerance range of component B may be set to + 0.05mm and the tolerance range of component C may be set to + 0.05mm, depending on the cost of components B and C. Namely, the component B with the tolerance range of +/-0.05 mm and the component C with the tolerance range of +/-0.05 mm are purchased, so that the mismatching of the sizes of the components B and C can be avoided, and the cost of incoming materials can be saved.

In some embodiments of the present disclosure, errors in the manufacturing process may result in assembly errors of the components, which may result in unacceptable assemblies. Fig. 5 shows a process flow for this case.

At block S502 of fig. 5, components of the assembly associated with features that result in failed assemblies are determined. Under the above assumption, by comparing corresponding images of a pass assembly and a fail assembly, it can be determined that the components of the assembly associated with the feature that caused the fail assembly to fail are component a, component B, component C, component D, component E, and component E'.

At block S504, a second one of the features associated with an assembly error of the component is determined. In the above examples, the components associated with the features that caused the reject assembly to reject are component a, component B, component C, component D, component E, and component E'. The second feature associated with assembly errors of the components among the features that result in unacceptable assembly failure may include misalignment of the placement angle of component a, misalignment of component E with component E', and misalignment of the placement position of component D with component C.

At block S506, a target process associated with the second feature of the plurality of specified processes is determined. In the above example, it can be seen with reference to FIG. 2 that the target processes associated with the second feature may include process I associated with misalignment of part A, process VII and process VIII associated with misalignment of part E and part E', and process IV and process V associated with misalignment of part D and part C.

At block S508, a tolerance range for the target process is determined from the images of the failed assembly and the qualified assembly that include the second feature. In the hypothetical example above, the placement angle of part A of a pass assembly and a fail assembly can be compared by the image at block 220. For example, the normal placement angle of component A would be such that the long side of component A would be parallel to the y-axis. In the actual assembly process, there is a possibility that the placement angle of the component a is deviated. The included angles between the long sides of the components A and the y axis in the qualified assembly are all within +/-2 degrees, and the included angles between the long sides of the components A and the y axis in the unqualified assembly are all over the range of +/-2 degrees. It can be determined that the long side of component a can be made to not affect the efficacy of the assembly when it is at an angle between ±2° to the y-axis. In other words, the tolerance range of process I for assembling component a may be ±2°. Therefore, when, for example, purchasing equipment for performing process I, equipment with a tolerance range of + -2 deg. should be selected.

Similarly to the manner in which the tolerance ranges of the devices performing process I, process IV, process V, process VII, and process VIII may be determined.

In the hypothetical example above, misalignment of component E with component E' results in assembly failure. In some embodiments of the present disclosure, the tolerance ranges of the apparatus performing process VII and the apparatus performing process VIII may be determined by considering the positions of the components E and E' in combination. Fig. 6 shows a process flow for this case.

At block S602, a set of associated features among the features that result in the failed assembly being failed is determined. In the above example, it may be determined that the dimensional match of part B to part C is a set of associated features (first set of associated features). The positional match of component E and component E' is a set of associated features (second set of associated features). The relative positions of component C and component D and the length of component D are a set of associated features (third set of associated features).

At block S604, a plurality of components in the assembly associated with the set of associated features is determined. In the hypothetical example above, it may be determined that the plurality of components associated with the first set of associated features are component B and component C, the plurality of components associated with the second set of associated features are component E and component E', and the plurality of components associated with the third set of associated features are component C and component D.

At block S606, a second feature of the set of associated features associated with assembly errors of the plurality of components is determined. In the above hypothetical example, the second feature associated with assembly errors of the plurality of components in the second set of associated features is the placement of component E with component E'. The second feature associated with assembly errors of the plurality of components in the third set of associated features is the placement of component C with component D. While the first set of associated features does not relate to the second features associated with assembly errors of the components.

At block S608, a plurality of target processes associated with the second feature among the plurality of specified processes is determined. In the above hypothetical example, the placement of the components E and E' is performed by the process VII and the process VIII. Thus, process VII and process VIII can be determined as target processes. The placement of the components C and D is performed by process IV and process V. Thus, process IV and process V can be determined as target processes.

At block S610, a tolerance range for each of the plurality of target processes is determined based on the images of the rejected and acceptable assemblies including the second feature and the cost of the equipment performing the plurality of target processes. In the hypothetical example above, the tolerance ranges for process VII and process VIII can be determined from the images of the pass assemblies and the fail assemblies at block 290 for the second set of associated features. It is assumed that if the tolerance range of the equipment performing process VII is ±4°, the tolerance range of the equipment performing process VIII needs to be ±1° so that the placement of the components E and E' does not affect the efficacy of the assembly. If the tolerance range of the apparatus performing process VII is + -3 deg., the tolerance range of the apparatus performing process VIII may be + -2 deg.. It is assumed that the equipment performing process VII with a tolerance range of ±4° is 10 tens of thousands cheaper than the equipment performing process VII with a tolerance range of ±3°, while the equipment performing process VIII with a tolerance range of ±1° is 20 tens of thousands of equipment performing process VIII with a tolerance range of ±2°. The tolerance range of the apparatus performing process VII may be set to ±3° and the tolerance range of the apparatus performing process VIII may be set to ±2° according to the costs of the apparatus performing process VII and the apparatus performing process VIII. Therefore, the arrangement of the components E and E' does not affect the efficacy of the assembly, and the cost of equipment can be saved.

The tolerance ranges of the device performing process IV and the device performing process V may be determined in a similar manner to the determination of the tolerance ranges of the device performing process VII and the device performing process VIII.

In some cases, the combination of dimensional errors and assembly errors of the components may result in unacceptable assemblies. Fig. 7 shows a process flow for this case.

At block S702, a set of associated features among the features that result in the failed assembly being failed is determined. In the above example, it may be determined that the dimensional match of part B to part C is a set of associated features (first set of associated features). The positional match of component E and component E' is a set of associated features (second set of associated features). The relative positions of component C and component D and the length of component D are a set of associated features (third set of associated features).

At block S704, a plurality of components in the assembly associated with the set of associated features is determined. In the hypothetical example above, it may be determined that the plurality of components associated with the first set of associated features are component B and component C, the plurality of components associated with the second set of associated features are component E and component E', and the plurality of components associated with the third set of associated features are component C and component D.

At block S706, a first feature of the set of associated features associated with dimensional errors of the plurality of components is determined. In the above hypothetical example, the first feature associated with the dimensional errors of the plurality of components in the first set of associated features is the width of the bottom protrusion of component B and the width of the groove of component C. The first feature associated with the dimensional error of the component is not referred to in the second set of associated features. The first feature of the third set of associated features that is associated with the dimensional error of the component is the length of the component D.

At block S708, a second feature of the set of associated features associated with assembly errors of the plurality of components is determined. In the above hypothetical example, the second feature associated with assembly errors of the plurality of components in the second set of associated features is the placement of component E with component E'. The second feature associated with assembly errors of the plurality of components in the third set of associated features is the placement of component C with component D. While the first set of associated features does not relate to the second features associated with assembly errors of the components.

At block S710, a plurality of target processes associated with the second feature among the plurality of specified processes is determined. In the hypothetical example above, the assembly of the components E and E' in the second set of associated features is performed by process VII and process VIII. The assembly of component C and component D in the third set of associated features is performed by process IV and process V.

At block S712, a tolerance range for each of the plurality of components and a tolerance range for each of the plurality of target processes is determined from the images of the reject assembly and the pass assembly including the first feature and the image including the second feature, the cost of the plurality of components, and the cost of the apparatus performing the plurality of target processes. In the hypothetical example above, the third set of associated features includes both the first feature and the second feature. For the third set of associated features, a tolerance range for the length of component D and tolerance ranges for process IV and process V may be determined from the images of the pass and fail assemblies at block 260. Specifically, if the tolerance range of the component D is small, it is costly, but the requirements for the tolerance ranges of the process IV and the process V are low. The cost of the equipment to perform process IV and process V is low. If the tolerance range of component D is large, it is low cost, but the requirements for the tolerance ranges of process IV and process V are high. The cost of the equipment to perform process IV and process V is high. The tolerance ranges for component D and process IV and process V may be determined by considering the cost of component D and the cost of the equipment performing process IV and process V. Thus, the qualification rate of the assembly piece can be ensured, and the comprehensive cost can be saved.

In some embodiments of the present disclosure, the dimensions of the components may also be selected by means of corresponding images of the reject assembly and the pass assembly. In which case the components of the assembly that are associated with the features that caused the failed assembly to fail may be determined. A third one of the features associated with the size of the component is then determined. Next, the component is sized according to the image of the failed assembly and the failed assembly including the third feature. For example, in the example of fig. 2, if it is not determined what the area of component a should be during the new product development phase, several candidate components a having different areas may be tried. The area of component a is then determined based on the yield of the assembly in these several cases.

By the processing method of the embodiment of the present disclosure, not only the tolerance range of the size of the component but also the size of the component can be determined.

Fig. 8 shows a schematic block diagram of a processing device 800 for an assembly according to an embodiment of the disclosure. As shown in fig. 8, the apparatus 800 may include a processor 810 and a memory 820 storing a computer program. The computer programs, when executed by the processor 810, enable the apparatus 800 to perform the steps of the method 100 as shown in fig. 1. In one example, apparatus 800 may be a computer device or a cloud computing node. The device 800 may acquire a set of images acquired for each assembly during assembly of the plurality of assemblies. Wherein the set of images includes images acquired for the assembly at the end of each of the plurality of designated processes. The apparatus 800 may determine a pass assembly and a fail assembly of the plurality of assemblies. The device 800 may then compare the corresponding images of the failed assembly to determine the features that caused the failed assembly to fail.

In some embodiments of the present disclosure, the device 800 may determine the component of the assembly associated with the feature. The apparatus 800 may determine a first one of the features associated with dimensional errors of the component. The apparatus 800 may determine a tolerance range for the component from the images of the failed assembly and the failed assembly including the first feature.

In some embodiments of the present disclosure, the apparatus 800 may determine a set of associated features of the features. The apparatus 800 may determine a plurality of components in the assembly associated with the set of associated features. The apparatus 800 may determine a first feature of the set of associated features that is associated with dimensional errors of the plurality of components. The apparatus 800 may determine a tolerance range for each of the plurality of components based on the images of the failed assembly and the failed assembly including the first feature and the cost of the plurality of components.

In some embodiments of the present disclosure, the device 800 may determine the component of the assembly associated with the feature. The apparatus 800 may determine a second one of the features associated with assembly errors of the components. The apparatus 800 may determine a target process of the plurality of specified processes associated with the second feature. The device 800 may determine a tolerance range for the target process from the images of the rejected assembly and the acceptable assembly including the second feature.

In some embodiments of the present disclosure, the apparatus 800 may determine a set of associated features of the features. The apparatus 800 may determine a plurality of components in the assembly associated with the set of associated features. The apparatus 800 may determine a second feature of the set of associated features that is associated with an assembly error of the plurality of components. The apparatus 800 may determine a plurality of target processes of the plurality of designated processes associated with the second feature. The apparatus 800 may determine a tolerance range for each of the plurality of target processes based on the images of the rejected and acceptable assemblies including the second feature and the cost of the device performing the plurality of target processes.

In some embodiments of the present disclosure, the apparatus 800 may determine a set of associated features of the features. The apparatus 800 may determine a plurality of components in the assembly associated with the set of associated features. The apparatus 800 may determine a first feature of the set of associated features that is associated with dimensional errors of the plurality of components. The apparatus 800 may determine a second feature of the set of associated features that is associated with an assembly error of the plurality of components. The apparatus 800 may determine a plurality of target processes of the plurality of designated processes associated with the second feature. The apparatus 800 may determine a tolerance range for each of the plurality of components and a tolerance range for each of the plurality of target processes based on the image of the failed assembly and the image of the failed assembly including the first feature and the image of the second feature, the cost of the plurality of components, and the cost of the device performing the plurality of target processes.

In some embodiments of the present disclosure, the device 800 may determine the component of the assembly associated with the feature. The apparatus 800 may determine a third one of the features associated with the size of the component. The device 800 may determine the size of the component from the images of the failed assembly and the failed assembly including the third feature.

In embodiments of the present disclosure, processor 810 may be, for example, a Central Processing Unit (CPU), a microprocessor, a Digital Signal Processor (DSP), a processor of a multi-core based processor architecture, or the like. Memory 820 may be any type of memory implemented using data storage technology including, but not limited to, random access memory, read only memory, semiconductor-based memory, flash memory, disk storage, and the like.

Furthermore, in embodiments of the present disclosure, apparatus 800 may also include an input device 830, such as a camera, keyboard, mouse, etc., for acquiring images and numbers of the assembly. In addition, the apparatus 800 may further include an output device 840, such as a display, for outputting the processing result.

In other embodiments of the present disclosure, there is also provided a computer readable storage medium storing a computer program, wherein the computer program is capable of implementing the steps of the method as shown in fig. 1 and 3 to 7 when being executed by a processor.

In summary, the processing method according to the embodiments of the present disclosure can determine the feature of the assembly failure by means of the images acquired for a plurality of designated processes, and select the components of the assembly and the apparatus performing the processes based on the feature. Further, embodiments of the present disclosure also contemplate a combination of features that result in failure of an assembly, and selecting components of the assembly and equipment performing the process based on the combination to increase the yield of the assembly and/or reduce the cost. The embodiment of the disclosure is also beneficial to improving the production efficiency and reducing the consumption of human resources in the actual production process.

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus and methods according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

As used herein and in the appended claims, the singular forms of words include the plural and vice versa, unless the context clearly dictates otherwise. Thus, when referring to the singular, the plural of the corresponding term is generally included. Similarly, the terms "comprising" and "including" are to be construed as being inclusive rather than exclusive. Likewise, the terms "comprising" and "or" should be interpreted as inclusive, unless such an interpretation is expressly prohibited herein. Where the term "example" is used herein, particularly when it follows a set of terms, the "example" is merely exemplary and illustrative and should not be considered exclusive or broad.

Further aspects and scope of applicability will become apparent from the description provided herein. It is to be understood that various aspects of the application may be implemented alone or in combination with one or more other aspects. It should also be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure.

While several embodiments of the present disclosure have been described in detail, it will be apparent to those skilled in the art that various modifications and variations can be made to the embodiments of the present disclosure without departing from the spirit and scope of the disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for processing an assembly, comprising:

Acquire a group of images captured for each assembly during the process of assembling a plurality of assemblies, wherein the group of images includes an image captured for the assembly at the end of each process in a plurality of designated processes and before the start of the next process, and an image captured for the operating table before the start of the assembly operation;

determining qualified assemblies and unqualified assemblies among the plurality of assemblies according to whether the function of each assembly is normal; and

An operation of comparing the unqualified assembly with the corresponding images of the qualified assembly to determine the features that cause the unqualified assembly to be unqualified is performed by multiple deep learning models, a first deep learning model among the multiple deep learning models is configured to compare the differences between a group of images captured for the same assembly to determine the region of interest in the images captured at the end of each process, and multiple second deep learning models among the multiple deep learning models are configured to compare the same region of interest in the corresponding images of the multiple assemblies for the multiple specified processes, respectively, to determine the features that cause the unqualified assembly to be unqualified.

2. The processing method according to claim 1, further comprising:

determining a component of the assembly associated with the feature;

determining a first one of the features that is associated with a dimensional error of the component;

A tolerance range of the component is determined based on the images of the unacceptable assembly and the acceptable assembly including the first feature.

3. The processing method according to claim 1, further comprising:

determining a set of associated features among the features;

determining a plurality of components in the assembly that are associated with the set of associated features;

determining a first feature in the set of associated features that is associated with a dimensional error of the plurality of parts;

A tolerance range for each of the plurality of components is determined based on the images of the unacceptable assembly and the acceptable assembly including the first feature and the costs of the plurality of components.

4. The processing method according to claim 1, further comprising:

determining a component of the assembly associated with the feature;

determining a second one of the features that is associated with an assembly error of the component;

determining a target process associated with the second feature among the plurality of designated processes;

The tolerance range of the target process is determined according to the images of the unqualified assembly and the qualified assembly including the second feature.

5. The processing method according to claim 1, further comprising:

determining a set of associated features among the features;

determining a second feature in the set of associated features that is associated with an assembly error of the plurality of components;

determining a plurality of target processes associated with the second feature from among the plurality of designated processes;

A tolerance range of each of the plurality of target processes is determined according to the images of the unqualified assembly and the qualified assembly including the second feature and the cost of equipment for performing the plurality of target processes.

6. The processing method according to claim 1, further comprising:

determining a set of associated features among the features;

A tolerance range of each of the plurality of components and a tolerance range of each of the plurality of target processes are determined based on the images including the first feature and the images including the second feature of the unqualified assembly and the qualified assembly, costs of the plurality of components, and costs of equipment that executes the plurality of target processes.

7. The processing method according to claim 1, further comprising:

determining a component of the assembly associated with the feature;

determining a third one of the features that is associated with a dimension of the component;

The size of the component is determined based on the images of the unacceptable assembly and the acceptable assembly including the third feature.

8 . The processing method according to claim 1 , wherein the plurality of second deep learning models are trained respectively through corresponding images of a plurality of test assemblies in which unqualified features and qualified features have been annotated.

9. A processing device for an assembly, comprising:

at least one processor; and

at least one memory storing a computer program;

Wherein, when the computer program is executed by the at least one processor, the processing device is caused to perform the steps of the processing method according to any one of claims 1 to 8.