AOI optical configuration method and system based on defect area image, electronic equipment, storage medium and computer program product

Abstract

The invention provides an AOI optical configuration method and system based on a defect area image, electronic equipment, a storage medium and a computer program product. The AOI optical configuration method comprises the following steps: acquiring defect area images of a preset number of to-be-detected samples, wherein the defect area images are determined in a configuration environment of initial AOI optical configuration; performing image differential processing on the defect area images of the preset number of to-be-detected samples and the standard image of the standard sample to determine the defect exposure capability of the initial AOI optical configuration; and determining a target AOI optical configuration which enables the defect exposure capability to meet a preset requirement according to a preset search algorithm, so as to indicate a user to adjust the initial optical configuration according to the target AOI optical configuration.

Description

Technical Field

[0001] The present invention relates to the technical field of automated optical inspection. Specifically, the present invention relates to an AOI optical configuration method, system, electronic device, storage medium, and computer program product based on an image of a defective area. Background Art

[0002] An AOI system (Automated Optical Inspection) is a defect detection system based on machine vision technology. For example, an AOI system can automatically identify defects on the surface of a sample to be detected (such as a wafer), such as solder joint errors or component missing defects, based on optical principles and image processing techniques.

[0003] In the design stage of an AOI system, a corresponding AOI optical configuration needs to be designed according to the physical properties of the sample to be detected, so that the defects on the surface of the sample to be detected can be clearly identified by the AOI system. However, the inventors of the present application have found that the current AOI optical configuration method of the AOI system relies relatively heavily on manual work. That is, it is necessary to manually evaluate the image quality of the AOI system and make corresponding adjustments to the AOI optical configuration according to the evaluation results.

[0004] The inventors have also found that, on the one hand, the evaluation results based on human subjective cognition often focus on the overall clarity of the image, and the actual severity of the defects of the sample to be detected may not be completely positively correlated with the degree of defects shown in the overall image, resulting in relatively poor accuracy of the evaluation results. On the other hand, the method of manually evaluating the image quality of the AOI system is difficult to objectively evaluate the image quality on a large data scale, cannot be combined with automated algorithms, and lacks intelligence.

[0005] The content of the background art section is only the technology known to the applicant and does not of course represent the prior art in this field. Summary of the Invention

[0006] According to an aspect of the present invention, the present invention provides an AOI optical configuration method based on an image of a defective area, including: obtaining images of defective areas of a preset number of samples to be detected, the images of defective areas being determined in a configuration environment of an initial AOI optical configuration; performing image difference processing on the images of defective areas of the preset number of samples to be detected and a standard image of a standard sample to determine the defect exposure ability of the initial AOI optical configuration; determining a target AOI optical configuration that satisfies a preset requirement according to a preset search algorithm to instruct a user to adjust the initial optical configuration according to the target AOI optical configuration.

[0007] According to another aspect of the present invention, the present invention provides an AOI optical configuration device based on an image of a defective area, including a conveying module, a supporting module, an image acquisition module, a first imaging light source module, a second imaging light source module, a light source control module, and a processing module. A preset number of placement points for samples to be detected are provided on the conveying module; the supporting module is arranged on the conveying module; the image acquisition module is arranged on the supporting module and is used to acquire images of the samples to be detected; the first imaging light source module emits first imaging light rays and is arranged on the supporting module; the second imaging light source module emits second imaging light rays; the second imaging light source module of the light source control module is arranged on the light source control module, and the light source control module is used to control the incident angle and intensity of the light source of the second imaging light rays; the processing module is electrically connected to the image acquisition module; wherein, the processing module acquires images of defective areas of a preset number of samples to be detected, and the images of defective areas are determined under the configuration environment of the initial AOI optical configuration. Image difference processing is performed on the images of defective areas of a preset number of samples to be detected and the standard images of standard samples to determine the defect exposure ability of the initial AOI optical configuration. According to a preset search algorithm, a target AOI optical configuration that enables the defect exposure ability to meet the preset requirements is determined to instruct the user to adjust the initial optical configuration according to the target AOI optical configuration.

[0008] According to another aspect of the present invention, the present invention also provides an electronic device. The electronic device includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors can implement the AOI optical configuration method as described above.

[0009] According to another aspect of the present invention, the present invention also provides a non-volatile computer-readable storage medium. A computer program is stored on the storage medium, and when the computer program is executed by a processor, it can implement the AOI optical configuration method as described above.

[0010] According to another aspect of the present invention, the present invention also provides a computer program product. The computer program product includes: a computer program stored on a computer-readable storage medium; the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is caused to execute the AOI optical configuration method as described above.

[0011] Beneficial effects

[0012] The present invention can determine the defect exposure ability of the defect area image by performing image difference processing on the defect area images of a preset number of samples to be detected and the standard images of the standard samples. And the present invention can optimize the AOI optical configuration at the algorithm adaptation level by determining the target AOI optical configuration that maximizes the defect exposure ability according to a preset search algorithm.

[0013] On the one hand, the present invention can achieve automatic optimization of the AOI optical configuration through machine learning, replacing the current method of manually adjusting the AOI optical configuration, which can avoid the influence of human subjective cognition and thus improve the accuracy of optimizing the AOI optical configuration. On the other hand, the present invention can objectively evaluate the image quality based on machine learning automation on a large data scale. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 A flowchart showing the AOI optical configuration method according to an embodiment of the present invention;

[0016] Figure 2 Another flowchart showing the AOI optical configuration method according to an embodiment of the present invention;

[0017] Figure 3 A schematic diagram showing the marked image of the defect area according to an embodiment of the present invention;

[0018] Figure 4 A schematic diagram showing the defect area image according to an embodiment of the present invention;

[0019] Figure 5 Another flowchart showing the AOI optical configuration method according to an embodiment of the present invention;

[0020] Figure 6 Another flowchart showing the AOI optical configuration method according to an embodiment of the present invention;

[0021] Figure 7 Another flowchart showing the AOI optical configuration method according to an embodiment of the present invention;

[0022] Figure 8 An iterative schematic diagram showing the search variables according to an embodiment of the present invention;

[0023] Figure 9Schematic structural diagram of the AOI optical configuration device showing an embodiment of the present invention;

[0024] Figure 10 Schematic diagram showing the control of the light source incident angle of an embodiment of the present invention.

[0025] Explanation of reference numerals:

[0026] AOI optical configuration device 10; conveying module 11; support module 12; image acquisition module 13; first imaging light source module 14; second imaging light source module 15; light source control module 16; processing module 17. Detailed implementation manners

[0027] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. Like reference numerals in the figures denote like or similar parts, and thus their repetitive description will be omitted.

[0028] The features, structures, or characteristics described may be combined in one or more embodiments in any suitable manner. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of these specific details, or in other ways, components, materials, devices, etc. In these cases, well-known structures, methods, devices, implementations, materials, or operations will not be shown or described in detail.

[0029] Furthermore, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0030] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order.

[0031] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0032] According to one aspect of the present invention, the present invention provides an AOI optical configuration method based on defect region images. Figure 1 The flowchart of the AOI optical configuration method according to an embodiment of the present invention is shown. As Figure 1 shown, the AOI optical configuration method may include steps S100 - S300.

[0033] Exemplarily, the AOI optical configuration method may be executed by an AOI optical configuration system with computing capabilities.

[0034] According to an exemplary embodiment, in step S100, the AOI optical configuration system acquires defect region images of a preset number of samples to be detected, and the defect region images are determined in the configuration environment of the initial AOI optical configuration.

[0035] For example, the preset number can be custom - set according to user requirements. The sample to be detected can be a target sample (such as a wafer) with a defect region on its surface selected by the user. The defect region image can be an image of the sample to be detected marked with the position of the defect region. The optical configuration includes at least the configuration of various configuration parameters of the AOI system. The initial AOI optical configuration can at least include the configuration of configuration parameters such as the incident angle of the light source, the intensity of the light source, and the speed of the moving platform, but the present invention is not limited thereto.

[0036] Exemplarily, the preset number can be 100, and the sample to be detected can be a wafer with a defect region on its surface. The AOI optical configuration system can acquire defect region images of 100 wafers with defect regions on their surfaces in the configuration environment of the initial AOI optical configuration.

[0037] Figure 2 Another flowchart of the AOI optical configuration method according to an embodiment of the present invention is shown; Figure 3 The schematic diagram of the defect region marked image according to an embodiment of the present invention is shown; Figure 4 The schematic diagram of the defect region image according to an embodiment of the present invention is shown; Figure 5 Another flowchart of the AOI optical configuration method according to an embodiment of the present invention is shown.

[0038] Optionally, as Figure 2 shown, step S100 may further include steps S110 - S113.

[0039] In step S110, the AOI optical configuration system acquires a defect region marked image of the sample to be detected under the first imaging light. The defect region on the surface of the sample to be detected is marked with a preset mark that can be recognized by the first imaging light.

[0040] For example, the preset marker can be a fluorescent marker pre - marked by the user on the defective area of the surface of the sample to be detected. The first imaging light can be UV light (ultraviolet light).

[0041] For example, as Figure 3 shown, since the fluorescent marker will be activated and emit visible light under the irradiation of UV light, the AOI optical configuration system can, under the irradiation of UV light, obtain a defective area marked image with the defective area marked (as Figure 3 shown by the circles in).

[0042] In step S111, the AOI optical configuration system obtains an initial image of the defective area of the sample to be detected under the second imaging light.

[0043] For example, the second imaging light can be visible light. The AOI optical configuration system can, under the irradiation of visible light, obtain an initial image of the defective area (i.e., the natural image of the sample to be detected without being processed under visible light).

[0044] In step S112, the AOI optical configuration system performs defective area annotation on the initial image of the defective area based on the defective area annotation image to obtain the defective area image of the sample to be detected.

[0045] For example, as Figure 4 shown, the AOI optical configuration system can, according to the defective area annotation image obtained under UV light and the initial image of the defective area obtained under visible light, after performing binarization processing on the defective area annotation image, obtain the defective area image under visible light in the initial image of the defective area.

[0046] In step S113, the AOI optical configuration system sequentially determines the defective area images of a preset number of samples to be detected.

[0047] For example, the AOI optical configuration system, based on the same method as above, sequentially obtains the defective area images of a preset number of samples to be detected (such as the defective area images of 100 wafers).

[0048] Through the above - mentioned embodiments, the present invention, through the marking method of fluorescent markers, performs image processing according to the defective area annotation image obtained under UV light and the initial image of the defective area obtained under visible light, so as to obtain a defective area image marked with the defective area.

[0049] The marking information formed by the fluorescent marker can change synchronously with the position of the image acquisition module or with image distortion. Therefore, in the case of image distortion caused by a change in the hardware environment, the present invention does not need to re - mark the defective area of the sample to be detected. The marking method of the sample to be detected provided by the present invention has the characteristic of high accuracy

[0050] Optionally, as Figure 2 shown, step S100 may further include step S114.

[0051] In step S114, the AOI optical configuration system performs image restoration on the defect area images of a preset number of samples to be detected based on a preset image restoration algorithm, so as to restore the interference items in the defect area images.

[0052] For example, the AOI optical configuration system may perform image restoration on the defect area images based on CoordFill (high-resolution image restoration). Exemplarily, the interference items may be ink marks in the defect area images, etc.

[0053] Through the above embodiments, the present invention can remove the interference items in the defect area images through image restoration of the defect area images, so that the image quality evaluation of the present invention is more accurate.

[0054] Optionally, as Figure 5 shown, step S100 may further include steps S120 - S122.

[0055] In step S120, the AOI optical configuration system acquires the initial defect area images of the samples to be detected under the second imaging light.

[0056] For example, the second imaging light may be visible light. The AOI optical configuration system may acquire the initial defect area images (i.e., the natural imaging of the samples to be detected without processing under visible light) under the irradiation of visible light.

[0057] In step S121, the AOI optical configuration system performs defect area annotation on the initial defect area images based on a preset image annotation tool to obtain the defect area images of the samples to be detected.

[0058] For example, the AOI optical configuration system may respond to a user instruction and perform defect area annotation on the initial defect area images through a preset image annotation tool, so as to obtain the defect area images marked with defect areas.

[0059] Exemplarily, the preset image annotation tool may be a data processing tool such as LabelMe (image annotation software), and the present invention is not limited thereto.

[0060] In step S122, the AOI optical configuration system sequentially determines the defect area images of a preset number of samples to be detected.

[0061] For example, the AOI optical configuration system acquires the defect area images of a preset number of samples to be detected (such as the defect area images of 100 wafers) based on the same method as above.

[0062] Through the above embodiments, the present invention can label the initial image of the defective area through a preset image annotation tool, and obtain a defective area image marked with the defective area.

[0063] In step S200, the AOI optical configuration system performs image difference processing on the defective area images of a preset number of samples to be detected and the standard images of the standard samples, so as to determine the defect exposure ability of the initial AOI optical configuration.

[0064] For example, image difference processing is a technique for detecting and analyzing image changes by comparing pixel differences between two or more images. The AOI optical configuration system can directly perform image difference processing on the defective area images of the preset number of samples to be detected and the standard images of the standard samples, or perform image difference processing after extracting image features. The present invention does not limit this.

[0065] Exemplarily, the image feature extraction methods used to extract image features include, but are not limited to, feature extraction based on a pre-trained neural network, and feature extraction based on feature extraction operators such as SIFT (Scale-Invariant Feature Transform) and HOG (Histogram of Oriented Gradients). The present invention does not limit this.

[0066] Through image difference processing, the present invention can compare the defective area marked in the defective area image with the intact area in the standard image, so as to evaluate the defect exposure ability under the initial AOI optical configuration according to the comparison result.

[0067] Figure 6 Another flowchart showing the AOI optical configuration method according to an embodiment of the present invention.

[0068] Optionally, as Figure 6 shown, step S200 may include steps S210 - S220.

[0069] In step S210, the AOI optical configuration system determines the standard area coordinate set and the defective area coordinate set in the defective area images of a preset number of samples to be detected.

[0070] In step S220, the AOI optical configuration system determines the defect exposure ability according to the standard area coordinate set and the defective area coordinate set.

[0071] For example, denote the preset number as n, denote the set of n samples to be detected of the same specification as D, and denote each sample to be detected as d, then:

[0072] d 0 ,d 1 ,d 2 ,...,d n ∈D;

[0073] Let the set of standard region coordinates in the defect region images of a preset number of samples to be detected be denoted as G, the set of defect region coordinates in the defect region images of a preset number of samples to be detected be denoted as NG, the ID of the sample d to be detected be denoted as i, and the standard sample be denoted as S. Then:

[0074]

[0075]

[0076] Among them, diff i is a variable used to describe the difference between image regions, and sim i is a variable used to describe the similarity between image regions. is the feature at the position of coordinates (x, y) on the sample i to be detected, and S x，y is the feature at the position of coordinates (x, y) on the standard sample S.

[0077] To make the features in the standard region as similar as possible, denote the defect exposure ability of each defect region image as dea i . Then:

[0078] dea i = diff i + sim i | diff i ∈ Diff i ;

[0079] Among them, Diff i is to maximize diff i . With this setting of the present invention, the recall of the defect region can be guaranteed as much as possible.

[0080] Denote the defect exposure ability of the initial AOI optical configuration as DEA. Then:

[0081]

[0082] Exemplarily, DEA (Defect Exposure Ability) can characterize the defect exposure ability under this initial AOI optical configuration. The larger DEA is, the stronger the defect exposure ability is characterized.

[0083] Optionally, the AOI optical configuration system can also determine DEA according to the image quality evaluation algorithm, such as LIQE (Image Quality Evaluation Method) and TOPIQ (Image Quality Evaluation Method Based on Attention Mechanism), etc. The present invention is not limited thereto.

[0084] In step S300, the AOI optical configuration system determines a target AOI optical configuration that enables the defect exposure ability to meet the preset requirements according to a preset search algorithm.

[0085] For example, the AOI optical configuration system determines multiple configuration parameters corresponding to the defect exposure ability that can meet the preset requirements in the search space formed by multiple configuration parameters in the optical configuration according to a preset search method, and thus determines the target AOI optical configuration according to the multiple configuration parameters.

[0086] Exemplarily, the preset search algorithm includes but is not limited to the coordinate descent method, the Newton descent method, the reinforcement learning algorithm, etc., and the present invention does not limit this.

[0087] Figure 7 Another flowchart showing the AOI optical configuration method according to an embodiment of the present invention.

[0088] Optionally, as Figure 7 shown, step S300 may include steps S310 - S340.

[0089] In step S310, the AOI optical configuration system determines at least two search variables of the target AOI optical configuration.

[0090] For example, the target AOI optical configuration of the AOI optical configuration system may include m search variables X (i.e., multiple configuration parameters in the target AOI optical configuration), such as X 1 , X 2 , X 3 ... X m etc. X 1 , X 2 , X 3 ... X m can form the search space of the target AOI optical configuration.

[0091] In step S320, the AOI optical configuration system determines one of the at least two search variables as the initial search variable and determines the remaining search variables as fixed quantities.

[0092] In step S330, the AOI optical configuration system iterates the initial search variable based on a preset iteration formula until the defect exposure ability meets the preset requirements, and determines the target search variable corresponding to the initial search variable.

[0093] For example, taking the coordinate descent method and the Newton descent method as examples, each time one search variable is selected as the initial search variable, and the other m - 1 search variables are determined as fixed quantities to optimize the defect exposure ability DEA, such as the optimization goal can be max(DEA(X)).

[0094] The AOI optical configuration system iterates the initial search variables based on a preset iteration formula until the defect exposure ability DEA reaches a preset threshold or no longer increases. And the AOI optical configuration system records the iterated search variables corresponding to the current time as the target search variables of the initial search variables.

[0095] Exemplarily, the preset iteration formula can be:

[0096]

[0097] Wherein, is the i-th optical scheme parameter at iteration round t, is the i-th optical scheme parameter at iteration round t + 1, is the DEA at iteration round t, is the partial derivative of the DEA at iteration round t with respect to the i-th variable in the optical scheme parameter x.

[0098] In step S340, the AOI optical configuration system traverses all search variables, determines the target search variables corresponding to all search variables, and determines the target AOI optical configuration according to the target search variables corresponding to all search variables.

[0099] For example, after determining the target search variables of the initial search variables, based on the same iteration method, the remaining m - 1 search variables are iterated in turn. Thus, the AOI optical configuration system can obtain the target search variables corresponding to each search variable. The AOI optical configuration system determines the target AOI optical configuration according to the target search variables corresponding to each search variable.

[0100] Through the above embodiments, the present invention can iteratively process the search variables (i.e., the configuration parameters in the target AOI optical configuration) automatically through machine learning, so as to determine multiple configuration parameters that can make the defect exposure ability meet the preset requirements.

[0101] In step S300, the AOI optical configuration system can instruct the user to adjust the initial optical configuration according to the target AOI optical configuration.

[0102] For example, after determining the multiple configuration parameters of the target AOI optical configuration, the AOI optical configuration system can instruct the user to adjust the initial optical configuration of the AOI optical configuration system. After that, the AOI optical configuration system can calculate the defect exposure ability of the adjusted optical configuration again until the defect exposure ability can meet the preset requirements (such as the defect exposure ability reaches the preset threshold or the defect exposure ability no longer changes, etc.).

[0103] Exemplarily, Figure 8Schematic diagram of iterative search variables according to an embodiment of the present invention. As Figure 8 shown, the present invention can perform iteration on multiple search variables (such as Figure 8 configuration parameters such as the light incident angle and light intensity shown in

[0104] ), and can determine the target light incident angle, target light intensity, etc. that enable the defect exposure ability to meet the preset requirements through the coordinate descent method and the Newton descent method.

[0105] On the one hand, the present invention can achieve automatic optimization of AOI optical configuration through machine learning, replacing the current method of adjusting AOI optical configuration manually, and can avoid the influence of human subjective cognition, thereby improving the accuracy of AOI optical configuration optimization. On the other hand, the present invention can objectively evaluate the image quality based on machine learning automation on a large data scale.

[0106] Figure 9 Schematic diagram of the structure of the AOI optical configuration device according to an embodiment of the present invention.

[0107] According to another aspect of the present invention, the present invention provides an AOI optical configuration device based on a defect area image. As Figure 9 shown, the AOI optical configuration device 10 may include a conveying module 11, a support module 12, an image acquisition module 13, a first imaging light source module 14, a second imaging light source module 15, a light source control module 16, and a processing module 17.

[0108] According to an exemplary embodiment, placement points for a preset number of samples to be detected are provided on the conveying module 11.

[0109] For example, multiple placement points for samples to be detected may be provided on the conveying module 11, and the conveying module 11 can move based on a preset moving platform speed, so as to achieve automatic transmission of the samples to be detected. Exemplarily, the conveying module 11 may be a conveyor belt.

[0110] The support module 12 is provided on the conveying module 11. Exemplarily, the support module 12 may be a support frame.

[0111] The image acquisition module 13 is provided on the support module 12 and is used to acquire images of the samples to be detected.

[0112] For example, as Figure 9 shown, the image acquisition module 13 is disposed above the image acquisition point of the sample to be detected. When the transfer module 11 uniformly drives the movement of the sample to be detected, the image acquisition module 13 can acquire an image of the sample to be detected at the image acquisition point and send the image to the processing module 17. Exemplarily, the image acquisition module 13 can be a line scan camera.

[0113] The first imaging light source module 14 can emit first imaging light, and the first imaging light source module 14 is disposed on the support module 12.

[0114] For example, the first imaging light can be UV light (ultraviolet light). The first imaging light source module 14 can provide UV light for the image acquisition point, so that the image acquisition module 13 can acquire an image under UV light.

[0115] The second imaging light source module 15 can emit second imaging light.

[0116] For example, the second imaging light can emit visible light. The second imaging light source module 15 can provide visible light for the image acquisition point, so that the image acquisition module 13 can acquire an image under visible light.

[0117] The second imaging light source module 15 is disposed on the light source control module 16, and the light source control module 16 is used to control the light incident angle and light intensity of the second imaging light.

[0118] Figure 10 The control schematic diagram of the light source incident angle according to the embodiment of the present invention is shown.

[0119] As Figure 10 shown, the present invention can adjust the position of the second imaging light source module 15 through the light source control module 16, and further can adjust the light source incident angle and light source intensity of the second imaging light emitted by the second imaging light source module 15. Exemplarily, the light source control module 16 can be a robotic arm, but the present invention is not limited thereto.

[0120] According to the exemplary embodiment, the processing module 17 is electrically connected to the image acquisition module 13. Exemplarily, the processing module 17 can be any electronic device with computing capabilities, including but not limited to a host, a server, or any intelligent processing terminal, etc.

[0121] According to an exemplary embodiment, the processing module 17 acquires defect area images of a preset number of samples to be detected, where the defect area images are determined in a configuration environment of an initial AOI optical configuration. Image differential processing is performed on the defect area images of the preset number of samples to be detected and the standard images of the standard samples to determine the defect exposure ability of the initial AOI optical configuration. A target AOI optical configuration that satisfies a preset requirement is determined according to a preset search algorithm, so as to instruct the user to adjust the initial optical configuration according to the target AOI optical configuration.

[0122] The processing module 17 is used to execute the AOI optical configuration method as described above, that is, the processing module 17 can be equivalent to the AOI optical configuration system as described above. Since the specific process of the AOI optical configuration method based on the defect area image has been described in detail above, it will not be elaborated here.

[0123] In the present invention, by performing image differential processing on the defect area images of a preset number of samples to be detected and the standard images of the standard samples, the defect exposure ability of the defect area images can be determined according to a preset algorithm. And in the present invention, by determining a target AOI optical configuration that maximizes the defect exposure ability according to a preset search algorithm, optimization of the AOI optical configuration can be achieved at the algorithm adaptation level.

[0124] On the one hand, the present invention can realize automatic optimization of the AOI optical configuration through machine learning, replacing the current method of adjusting the AOI optical configuration manually, which can avoid the influence of human subjective cognition, thereby improving the accuracy of optimizing the AOI optical configuration. On the other hand, the present invention can objectively evaluate the image quality based on machine learning automation on a large data scale.

[0125] According to another aspect of the present invention, the present invention also provides an electronic device. The electronic device includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors can implement the AOI optical configuration method as described above.

[0126] According to another aspect of the present invention, the present invention also provides a non-volatile computer-readable storage medium. A computer program is stored on the storage medium, and when the computer program is executed by a processor, it can implement the AOI optical configuration method as described above.

[0127] According to another aspect of the present invention, the present invention also provides a computer program product. The computer program product includes: a computer program stored on a computer-readable storage medium; the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is caused to execute the AOI optical configuration method as described above.

[0128] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions of the foregoing embodiments or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An AOI optical configuration method based on defect area image, characterized in that: include: Acquire a preset number of defect area images of samples to be inspected, wherein the defect area images are determined under a configuration environment of an initial AOI optical configuration; Performing image difference processing on the defect area images of the preset number of samples to be inspected and the standard image of the standard sample to determine the defect exposure capability of the initial AOI optical configuration; A target AOI optical configuration that enables the defect exposure capability to meet preset requirements is determined according to a preset search algorithm, so as to instruct the user to adjust the initial optical configuration according to the target AOI optical configuration.

2. The AOI optical configuration method according to claim 1, characterized in that: The step of obtaining a preset number of defect area images of samples to be inspected comprises: Acquire a defect region marking image of the sample to be detected under a first imaging light, wherein the defect region on the surface of the sample to be detected is marked with a preset mark that can be recognized by the first imaging light; Acquire an initial image of a defect area of ​​the sample to be inspected under a second imaging light; Annotating the defect area of ​​the initial defect area image based on the defect area annotation image to obtain the defect area image of the sample to be detected; The defect area images of the preset number of samples to be inspected are determined in sequence.

3. The AOI optical configuration method according to claim 1, characterized in that: The step of obtaining a preset number of defect area images of samples to be inspected comprises: Acquire an initial image of a defect area of ​​the sample to be inspected under a second imaging light; Annotating the defect area of ​​the initial defect area image based on a preset image annotation tool to obtain the defect area image of the sample to be detected; The defect area images of the preset number of samples to be inspected are determined in sequence.

4. The AOI optical configuration method according to claim 2, characterized in that: After sequentially determining the defect area images of the preset number of samples to be inspected, the AOI optical configuration method further includes: Image restoration is performed on the defect area images of the preset number of samples to be detected based on a preset image restoration algorithm to restore interference items in the defect area images.

5. The AOI optical configuration method according to claim 1, characterized in that: The performing image difference processing on the defect area images of the preset number of samples to be inspected and the standard images of the standard samples to determine the defect exposure capability of the initial AOI optical configuration includes: Determining a standard region coordinate set and a defect region coordinate set in defect region images of the preset number of samples to be inspected; The defect exposure capability is determined according to the standard region coordinate set and the defect region coordinate set.

6. The AOI optical configuration method according to claim 1, characterized in that: Determining the target AOI optical configuration that enables the defect exposure capability to meet the preset requirements according to the preset search algorithm includes: determining at least two search variables for the target AOI optical configuration; Determine one of the at least two search variables as an initial search variable and determine the remaining search variables as quantitative; Iterating the initial search variable based on a preset iterative formula until the defect exposure capability meets a preset requirement, and determining a target search variable corresponding to the initial search variable; All search variables are traversed to determine target search variables corresponding to all the search variables, so as to determine the target AOI optical configuration according to the target search variables corresponding to all the search variables.

7. An AOI optical configuration device based on defect area image, characterized in that: Used to perform the AOI optical configuration method according to any one of claims 1 to 6, the AOI optical configuration device comprising: A conveying module, wherein a preset number of placement points for samples to be tested are arranged on the conveying module; A supporting module, arranged on the conveying module; An image acquisition module, arranged on the support module, for acquiring an image of the sample to be detected; A first imaging light source module, disposed on the supporting module, and configured to emit a first imaging light; A second imaging light source module, used for emitting a second imaging light; A light source control module, on which the second imaging light source module is arranged, and the light source control module is used to control the light source incident angle and light source intensity of the second imaging light; A processing module, electrically connected to the image acquisition module; The processing module obtains defect area images of a preset number of samples to be inspected, the defect area images are determined under the configuration environment of the initial AOI optical configuration, performs image difference processing on the defect area images of the preset number of samples to be inspected and the standard images of the standard samples to determine the defect exposure capability of the initial AOI optical configuration, determines a target AOI optical configuration that makes the defect exposure capability meet preset requirements according to a preset search algorithm, and instructs the user to adjust the initial optical configuration according to the target AOI optical configuration.

8. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the AOI optical configuration method according to any one of claims 1 to 6.

9. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the AOI optical configuration method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The method comprises a computer program stored on a computer-readable storage medium, wherein the computer program comprises program instructions. When the program instructions are executed by a computer, the computer is enabled to execute the AOI optical configuration method according to any one of claims 1 to 6.

Images