Defect detection method and device, electronic device

Abstract

The application provides a defect detection method and device and electronic equipment. The defect detection method comprises the following steps: acquiring gray scale data of a display panel to be detected; performing feature extraction on the gray scale data based on a target defect to be detected, to obtain target feature data; determining a target defect condition corresponding to the target defect from a pre-stored corresponding relationship between defects and defect conditions; detecting whether there is data in the target feature data satisfying the target defect condition; and if there is, determining that the display panel has the target defect. In this way, direct, rapid and accurate detection of the target defect in the display panel is realized, and the detection efficiency is effectively improved.

Description

Technical Field

[0001] This application relates to the field of defect detection technology, specifically to a defect detection method and apparatus, and electronic equipment. Background Technology

[0002] Currently, display panel production is characterized by long processes, numerous production sites, and complex technologies. To ensure the yield rate of display panels, defect detection is required after each step of the production process to prevent defective products from entering subsequent processes and causing resource waste.

[0003] There are hundreds of types of defects generated during the production of display panels. Among these, only a small percentage are lethal defects that severely affect product quality. However, existing defect detection methods often require screening for lethal defects only after the inspection is completed, resulting in low defect detection efficiency. Summary of the Invention

[0004] In view of this, this application aims to provide a defect detection method, apparatus, and electronic device that can effectively improve defect detection efficiency.

[0005] The first aspect of this application provides a defect detection method, comprising:

[0006] Obtain the grayscale data of the display panel to be inspected;

[0007] Based on the target defect to be detected, feature extraction is performed on the grayscale data to obtain target feature data;

[0008] From the pre-stored correspondence between defects and defect conditions, determine the target defect condition corresponding to the target defect;

[0009] The system detects whether any data in the target feature data satisfies the target defect condition; if so, it determines that the display panel has a target defect.

[0010] Optionally, the step of extracting features from the grayscale data based on the target defect to be detected to obtain target feature data includes:

[0011] Based on the pre-stored correspondence between defects and feature items, the target feature item corresponding to the target defect is determined;

[0012] Based on the target feature terms, feature extraction is performed on the grayscale data to obtain the target feature data.

[0013] Optionally, the method for constructing the correspondence between defects and feature terms includes:

[0014] Acquire multiple sampling data corresponding to the defect; the sampling data includes multiple parameter items;

[0015] Among the multiple parameter items, the item that satisfies the correlation condition is taken as the feature item corresponding to the defect, and the correspondence between the defect and the feature item is obtained; the correlation condition includes: in the current parameter item, the proportion of the specified feature data in the item satisfies a first threshold.

[0016] Optionally, the items that satisfy the correlation condition among the plurality of parameter items are taken as feature items corresponding to the defect, and the correspondence between the defect and the feature items is obtained, including:

[0017] Among the plurality of parameter items, the item that satisfies the correlation condition is determined as the first relevant item;

[0018] The preset experience items and the first related items are deduplicated and merged to obtain the first merged item. The first merged item is used as the feature item corresponding to the defect to obtain the correspondence between the defect and the feature item.

[0019] Optionally, the step of using the first merged item as a feature item corresponding to the defect to obtain the correspondence between the defect and the feature item includes:

[0020] Determine the correlation value between each item in the target item and each item in the first merged item;

[0021] The items whose correlation values ​​among the target items meet the second threshold are identified as the second relevant items;

[0022] The second related item and the first merged item are merged to obtain the second merged item;

[0023] The second merged item is used as the feature item corresponding to the defect to obtain the correspondence between the defect and the feature item.

[0024] Optionally, the method for constructing the correspondence between defects and defect conditions includes:

[0025] Obtain a sample dataset and label the defects in the sample data within the sample dataset;

[0026] Based on the correspondence between the defects and the feature items, defects and their corresponding feature data are selected from the sample dataset;

[0027] The selected defects and their corresponding feature data are used as training data to train the random forest classification model, thereby obtaining the correspondence between the defects and the defect conditions.

[0028] Optionally, acquiring the grayscale data of the display panel to be detected includes:

[0029] Obtain the initial grayscale data of the display panel to be inspected;

[0030] The initial grayscale data is subjected to threshold processing to obtain the grayscale data.

[0031] Optionally, after determining that the display panel has a target defect, the method further includes:

[0032] Obtain the location information of the target defect from the data that meets the target defect conditions;

[0033] Based on the location information of the target defect, an image of the target defect is obtained.

[0034] A second aspect of this application provides a defect detection device, comprising:

[0035] The first acquisition module is used to acquire grayscale data of the display panel to be detected;

[0036] The feature extraction module is used to extract features from the grayscale data based on the target defect to be detected, so as to obtain target feature data;

[0037] The determination module is used to determine the target defect condition corresponding to the target defect from a pre-stored correspondence between defects and defect conditions;

[0038] The detection module is used to detect whether there is data in the target feature data that meets the target defect conditions; if so, it is determined that the display panel has a target defect.

[0039] A third aspect of this application provides an electronic device, including: a processor and a memory connected to the processor;

[0040] The memory is used to store computer programs;

[0041] The processor is used to call and execute the computer program in the memory to perform the defect detection method as described in the first aspect of this application.

[0042] This application provides a defect detection method, apparatus, and electronic device. The defect detection method pre-stores the correspondence between defects and defect conditions, providing different defect conditions based on different defects, thus offering an effective basis for defect screening of display panels. Based on this, after acquiring the grayscale data of the display panel to be inspected, feature extraction can be performed on the grayscale data based on the target defect to be detected, obtaining target feature data. Then, from the correspondence between defects and defect conditions, the target defect conditions corresponding to the target defect are determined, and it is further checked whether data in the target feature data satisfies the target defect conditions. If so, it can be determined that the display panel has a target defect. In this way, direct, rapid, and accurate detection of target defects in display panels is achieved, effectively improving detection efficiency. Attached Figure Description

[0043] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0044] Figure 1 This is a flowchart of a defect detection method provided in one embodiment of this application.

[0045] Figure 2 This is a schematic diagram of the structure of a defect detection device provided in one embodiment of this application.

[0046] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] Currently, due to factors such as long manufacturing processes, numerous production sites, and complex technologies, display panels generate hundreds of different types of defects during production. To avoid insufficient product yield, defect detection must be performed on products before proceeding to the next step in the production process. Unqualified products are discarded, and products with repairable defects (including potentially fatal defects) are repaired promptly. If product defects are not detected in time, they will result in mass scrapping and significant resource waste if they flow into subsequent processes.

[0049] Traditional automated optical inspection (AOI) machines and lamp-lighting machines can detect a massive number of defects during the display panel manufacturing process. However, not every defect affects the quality of the display panel. Therefore, after obtaining a large number of defects, it is necessary to screen out those that will affect the quality of the display panel. Taking lethal defects as an example, after detecting a large number of defects, it is still necessary to manually screen out the lethal defects, which is a cumbersome and time-consuming process, resulting in low inspection efficiency.

[0050] In view of this, embodiments of this application provide a defect detection method, such as... Figure 1 As shown, the method includes at least the following implementation steps:

[0051] S101. Obtain the grayscale data of the display panel to be tested.

[0052] In application, grayscale data of the display panel to be inspected can be obtained through AOI or charge-coupled device, providing basic data for defect detection.

[0053] S102. Based on the target defect to be detected, feature extraction is performed on the grayscale data to obtain target feature data.

[0054] The target defect can be one type of defect that the inspector needs to detect, or it can be multiple types of defects that the inspector needs to detect.

[0055] After obtaining the grayscale data of the display panel to be inspected, the target feature data can be extracted from the grayscale data based on the target defect, providing data support for subsequent target defect detection.

[0056] It is important to understand that target feature data refers to feature data related to the target defect. Extracting target feature data from grayscale data can avoid interference from invalid data. While improving the accuracy of detection, the reduction in data volume also improves detection efficiency.

[0057] S103. Determine the target defect condition corresponding to the target defect from the pre-stored correspondence between defects and defect conditions.

[0058] Since display panels can have various defects during the production process, in order to ensure the accuracy of defect detection, a correspondence between defects and defect conditions can be established in advance.

[0059] Specifically, the correspondence between defects and defect conditions can be set according to actual needs. It's important to note that different defects will have different defect conditions. For example, lethal defects in display panels include metal residue and small black spots; the criteria for judging these two are different, meaning the conditions for metal residue and small black spot phenomena will differ.

[0060] By pre-stored the correspondence between defects and defect conditions, the target defect conditions corresponding to the target defect can be determined, providing a basis for judging whether the display panel has a target defect. In this way, the rapid detection of specified defects can be achieved, improving detection efficiency while ensuring detection accuracy.

[0061] S104. Detect whether there is data in the target feature data that meets the target defect conditions; if so, determine that there is a target defect in the display panel.

[0062] If the target feature data contains data that satisfies the target defect condition, it indicates that there is abnormal data related to the target defect in the grayscale data. This abnormal data reflects the target defect, that is, the display panel has a target defect. Conversely, if there is no abnormal data related to the target defect in the grayscale data, it can be determined that the display panel does not meet the target defect condition.

[0063] For example, if the target defect is metal residue, metal residue feature data can be extracted from the acquired grayscale data. Then, based on the determined metal residue conditions, the metal residue feature data can be detected. If any data in the metal residue feature data meets the metal residue conditions, it can be determined that the display panel has a metal residue defect; otherwise, it can be determined that the display panel does not have a metal residue defect.

[0064] In this embodiment, a pre-stored correspondence between defects and defect conditions is used to provide different defect conditions based on different defects, providing an effective basis for defect screening of the display panel. Based on this, after acquiring the grayscale data of the display panel to be inspected, feature extraction can be performed on the grayscale data based on the target defect to be detected, obtaining target feature data. Then, from the correspondence between defects and defect conditions, the target defect conditions corresponding to the target defect are determined, and it is then checked whether any data in the target feature data satisfies the target defect conditions. If so, it can be determined that the display panel has a target defect. Thus, direct, rapid, and accurate detection of target defects in the display panel is achieved, effectively improving detection efficiency.

[0065] In some implementations, to ensure the accuracy of the detection results, when extracting features from grayscale data based on the target defect to be detected to obtain target feature data, the target feature item corresponding to the target defect can be determined based on the pre-stored correspondence between defects and feature items; and the target feature data can be obtained by extracting features from grayscale data according to the target feature item.

[0066] Based on the target feature terms, feature extraction is performed on the grayscale data. That is, the data corresponding to the target feature terms is extracted from the data under each parameter of the grayscale data, and used as the target feature data.

[0067] The feature items can be parameters related to the corresponding defect detection, selected from grayscale sampling data. Each feature item corresponding to a defect includes multiple parameter items. Correspondingly, different defects will have different corresponding feature items. By establishing the correspondence between defects and feature items, a basis can be provided for subsequent feature data extraction. This ensures that the extracted feature data meets the detection requirements while more accurately eliminating invalid data, thus improving the accuracy and efficiency of detection.

[0068] In some implementations, the method for constructing the correspondence between defects and feature items may include: acquiring multiple sampled data corresponding to the defect; the sampled data includes multiple parameter items; and selecting the item that satisfies the correlation condition from among the multiple parameter items as the feature item corresponding to the defect, thereby obtaining the correspondence between the defect and the feature item. The correlation condition may include: in the current parameter item, the proportion of a specified feature data in that item satisfies a first threshold.

[0069] The percentage meeting the first threshold can be either greater than or equal to the first threshold; conversely, the percentage not meeting the first threshold can be either less than the first threshold. In implementation, the first threshold can be set according to actual needs and is not limited here.

[0070] In application, the specified feature data can be a single data point belonging to a specified defect. For example, if the first threshold is 75%, and the specified feature data is data Y belonging to the metal residue defect, and there are 100 data points in the "Current Layer Defect" parameter, of which 76 belong to the metal residue defect data Y, and the remaining 24 are other data points, it can be seen that the proportion of data belonging to the metal residue defect in the "Current Layer Defect" parameter is 76%, exceeding the first threshold of 75%. In other words, the proportion of data belonging to the metal residue defect in the "Current Layer Defect" parameter meets the first threshold, and therefore the "Current Layer Defect" parameter can be added to the feature items corresponding to the metal residue defect.

[0071] Of course, the specified feature data can also be a segment of data belonging to a specified defect. For example, for data with a discrete distribution in the parameter item, the discretely distributed data can be divided into segments first. Then, for the current parameter item, whether the proportion of a segment of data belonging to the specified defect in that item meets a first threshold determines whether to add the current parameter item to the feature item corresponding to the specified defect. For example, if the first threshold is 70%, and the specified feature data is the 55-83 data segment belonging to the metal residue defect, there are 200 data points in the grayscale value parameter item. Among them, the number of data points belonging to the 55-83 data segment of the metal residue defect is 150, and the number of other data points is 50. It can be seen that the proportion of the 55-83 data segment belonging to the metal residue defect in the grayscale value parameter item is 75%, which exceeds the first threshold of 70%. That is, the proportion of the 55-83 data segment belonging to the metal residue defect in the grayscale value parameter item meets the first threshold, and the grayscale value parameter item can be added to the feature item corresponding to the metal residue defect.

[0072] In some implementations, when using the items that satisfy the correlation condition among multiple parameter items as feature items corresponding to the defect to obtain the correspondence between the defect and the feature items, the items that satisfy the correlation condition among multiple parameter items can also be determined as the first related item; the preset empirical items and the first related items are deduplicated and merged to obtain the first merged item, and the first merged item is used as the feature item corresponding to the defect to obtain the correspondence between the defect and the feature item.

[0073] The preset experience items can be set according to actual experience needs, and there are no restrictions here.

[0074] Adding preset experience items determined based on actual experience requirements to the feature items can fill in the gaps in the first relevant items selected by the correlation conditions, improve the relevance range of the feature items corresponding to the defects, avoid the loss of defect-related data, and thus improve the accuracy of defect detection.

[0075] In some implementations, when using the first merged item as a feature item corresponding to the defect to obtain the correspondence between the defect and the feature item, the correlation value between each item in the target item and each item in the first merged item can be determined first; then, the items in the target item whose correlation value meets the second threshold are determined as the second related items; the second related items and the first merged item are merged to obtain the second merged item; the second merged item is used as a feature item corresponding to the defect to obtain the correspondence between the defect and the feature item.

[0076] The second threshold can be set according to actual needs, and is not limited here.

[0077] In practice, the correlation value can be the Pearson correlation coefficient. Specifically, the calculation of the Pearson correlation coefficient between each item can be found in existing related techniques, and will not be elaborated upon here.

[0078] By merging the second relevant term and the first merged term, and using the second merged term as the feature term corresponding to the defect, the feature column corresponding to the defect can be further improved, thereby further improving the accuracy of defect detection.

[0079] In some implementations, the method for constructing the correspondence between defects and defect conditions may include the following steps: obtaining a sample dataset and labeling the defects present in the sample data in the sample dataset; then, based on the correspondence between defects and feature items, selecting defects and their corresponding feature data from the sample dataset; using the selected defects and their corresponding feature data as training data to train a random forest classification model to obtain the correspondence between defects and defect conditions.

[0080] The specific training process can be found in existing related technologies, and will not be detailed here.

[0081] Training a random forest classification model using training data, combining deep learning and random forest classification algorithms, yields efficient classification logic. Compared to traditional manually set classification logic, machine learning effectively improves classification accuracy. Furthermore, using feature data corresponding to defects as training data further enhances the accuracy of the classification logic, providing a more accurate reference for defect detection. In addition, the random forest algorithm allows for visualization of the classification logic, facilitating user viewing, extraction, and refinement, thus providing greater convenience for defect detection.

[0082] Of course, this application is not limited to using the random forest classification algorithm as the learning algorithm. In some other implementations, other classification algorithms may also be used, such as logistic regression, support vector machine, K-nearest neighbor algorithm, etc.

[0083] In some implementations, when acquiring grayscale data of the display panel to be inspected, the initial grayscale data of the display panel to be inspected can be acquired; the initial grayscale data can be thresholded to obtain the grayscale data.

[0084] When acquiring the initial grayscale data of the display panel to be inspected, AOI or charge-coupled device (CCD) methods can be used. This initial grayscale data is different from image data or reported data, resulting in more accurate and reliable data. After obtaining the initial grayscale data, thresholding can be performed to filter out some abnormal data, improving detection efficiency.

[0085] In some implementations, after determining that a target defect exists in the display panel, the defect detection method may further include: obtaining the location information of the target defect from data that meets the target defect conditions; and obtaining an image of the target defect based on the location information of the target defect.

[0086] It is important to understand that feature data contains location information. Similarly, data that meets the target defect conditions also carries corresponding location information. Data that meets the target defect conditions is the data that reflects the target defect. Therefore, the location information carried in the data that meets the target defect conditions is the location information of the target defect.

[0087] Based on the location information of the target defect, an image of the target defect is obtained, achieving the goal of obtaining an image of the defect, which helps to improve the product repair yield and enhance the assessment of product health.

[0088] Furthermore, unlike existing methods that acquire images of all defects in the display panel to be inspected, the present application's solution completes the classification and detection of defects before acquiring images, and then acquires the corresponding image for the target defect. This avoids the acquisition of images of non-target defects, effectively saving resources, and also avoids interference from non-target defect images, saving the process of manually screening target defects. This achieves rapid and accurate acquisition of images of target defects and improves detection efficiency.

[0089] As an optional implementation, embodiments of this application also provide a defect detection device, such as... Figure 2 As shown, the defect detection device may include: a first acquisition module 201, used to acquire grayscale data of the display panel to be detected; a feature extraction module 202, used to extract features from the grayscale data based on the target defect to be detected, to obtain target feature data; a determination module 203, used to determine the target defect condition corresponding to the target defect from a pre-stored correspondence between defects and defect conditions; and a detection module 204, used to detect whether there is data in the target feature data that satisfies the target defect condition; if so, it is determined that the display panel has a target defect.

[0090] Optionally, the feature extraction module 202 can be used to: determine the target feature item corresponding to the target defect based on the pre-stored correspondence between defects and feature items; and extract features from the grayscale data according to the target feature item to obtain the target feature data.

[0091] Optionally, the defect detection device may further include a first construction module, which is used to: acquire multiple sampling data corresponding to the defect; the sampling data includes multiple parameter items; among the multiple parameter items, the item that satisfies the correlation condition is taken as the feature item corresponding to the defect, and the correspondence between the defect and the feature item is obtained; the correlation condition includes: in the current parameter item, the proportion of the specified feature data in the item satisfies a first threshold.

[0092] Optionally, when using the items that satisfy the correlation condition among multiple parameter items as feature items corresponding to the defect to obtain the correspondence between the defect and the feature items, the first construction module can be used to: determine the items that satisfy the correlation condition among multiple parameter items as the first related items; perform deduplication and merging on the preset experience items and the first related items to obtain the first merged item; use the first merged item as the feature item corresponding to the defect to obtain the correspondence between the defect and the feature items.

[0093] Optionally, when using the first merged item as a feature item corresponding to the defect to obtain the correspondence between the defect and the feature item, the first construction module can be specifically used to: determine the correlation value between each item in the target item and each item in the first merged item; determine the items in the target item whose correlation value meets the second threshold as second related items; merge the second related items and the first merged item to obtain the second merged item; and use the second merged item as a feature item corresponding to the defect to obtain the correspondence between the defect and the feature item.

[0094] Optionally, the defect detection device may further include a second construction module, which may be used to: acquire a sample dataset and label the defects in the sample data in the sample dataset; based on the correspondence between defects and feature items, select defects and their corresponding feature data from the sample dataset; use the selected defects and their corresponding feature data as training data to train a random forest classification model to obtain the correspondence between defects and defect conditions.

[0095] Optionally, the first acquisition module can be used to: acquire the initial grayscale data of the display panel to be detected; and perform threshold processing on the initial grayscale data to obtain grayscale data.

[0096] Optionally, the defect detection device may further include a second acquisition module, which may be used to: acquire the location information of the target defect from data that meets the target defect conditions; and acquire an image of the target defect based on the location information of the target defect.

[0097] It should be understood that the specific implementation of the defect detection device provided in the embodiments of this application can refer to the specific implementation of the defect detection method described in any of the above embodiments, and will not be repeated here.

[0098] As an optional implementation, embodiments of this application also provide an electronic device, such as... Figure 3 As shown, the electronic device may include: a memory 301 and a processor 302; wherein, the memory 301 is connected to the processor 302 and is used to store a program; the processor 302 is used to implement the defect detection method disclosed in one embodiment above by running the program stored in the memory 301.

[0099] Specifically, the electronic device may also include: a bus, a communication interface 303, an input device 304, and an output device 305.

[0100] The processor 302, memory 301, communication interface 303, input device 304, and output device 305 are interconnected via a bus. Among them:

[0101] A bus can include a pathway for transmitting information between various components of a computer system.

[0102] Processor 302 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0103] Processor 302 may include a main processor, as well as a baseband chip, modem, etc.

[0104] The memory 301 stores a program that executes the technical solution of the present invention, and may also store an operating system and other key business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory 301 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0105] Input device 304 may include a device for receiving user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor.

[0106] Output device 305 may include devices that allow information to be output to a user, such as a display screen, a speaker, etc.

[0107] The communication interface 303 may include a device that uses any transceiver to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0108] The processor 302 executes the program stored in the memory 301, which can be used to implement the various steps of the defect detection method provided in the embodiments of this application.

[0109] In application, the electronic device can be AOI. The detection chamber of AOI does not affect the product yield. It can detect defects and capture defect locations without adding other equipment, thus reducing the product life cycle.

[0110] As an optional implementation, another embodiment of this application also provides a storage medium storing a computer program, which, when executed by a processor, implements the various steps of the defect detection method provided in any of the above embodiments.

[0111] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0112] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0113] The steps in the methods of the various embodiments of this application can be adjusted, combined, or deleted according to actual needs.

[0114] The modules and sub-modules in the apparatus and terminal in the various embodiments of this application can be merged, divided, and deleted according to actual needs.

[0115] It should be understood that the disclosed terminals, devices, and methods can be implemented in other ways, given the several embodiments provided in this application. For example, the terminal embodiments described above are merely illustrative. For instance, the division of modules or sub-modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple sub-modules or modules may be combined or integrated into another module, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0116] The modules or submodules described as separate components may or may not be physically separate. The components that constitute a module or submodule may or may not be physical modules or submodules; that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0117] Furthermore, the functional modules or sub-modules in the various embodiments of this application can be integrated into one processing module, or each module or sub-module can exist physically separately, or two or more modules or sub-modules can be integrated into one module. The integrated modules or sub-modules described above can be implemented in hardware or in the form of software functional modules or sub-modules.

[0118] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software unit executed by a processor, or a combination of both. The software unit can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0120] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0121] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A defect detection method, characterized in that, include: Obtain the grayscale data of the display panel to be inspected; Based on the target defect to be detected, feature extraction is performed on the grayscale data to obtain target feature data; From the pre-stored correspondence between defects and defect conditions, determine the target defect condition corresponding to the target defect; Detect whether there is any data in the target feature data that satisfies the target defect condition; if so, determine that the display panel has a target defect; The step of extracting features from the grayscale data based on the target defect to be detected, to obtain target feature data, includes: Based on the pre-stored correspondence between defects and feature terms, the target feature terms corresponding to the target defect are determined; Based on the target feature terms, feature extraction is performed on the grayscale data to obtain target feature data; The method for constructing the correspondence between defects and feature terms includes: Acquire multiple sampling data corresponding to the defect; the sampling data includes multiple parameter items; Among the multiple parameter items, the item that satisfies the correlation condition is taken as the feature item corresponding to the defect, and the correspondence between the defect and the feature item is obtained; the correlation condition includes: in the current parameter item, the proportion of the specified feature data in the item satisfies a first threshold; The step of selecting the item that satisfies the correlation condition from among the plurality of parameter items as the feature item corresponding to the defect includes: When the data in the parameter item is discretely distributed, the parameter item is segmented to obtain multiple candidate data segments; Determine the proportion of the data belonging to the specified defect in each of the candidate data segments; The parameter items corresponding to the candidate data segments whose proportions meet the first threshold are determined as the feature items corresponding to the defect.

2. The method according to claim 1, characterized in that, Among the multiple parameter items, those that satisfy the correlation condition are taken as feature items corresponding to the defect, thus obtaining the correspondence between the defect and the feature items, including: Among the plurality of parameter items, the item that satisfies the correlation condition is determined as the first relevant item; The preset experience items and the first related items are deduplicated and merged to obtain the first merged item. The first merged item is used as the feature item corresponding to the defect to obtain the correspondence between the defect and the feature item.

3. The method according to claim 1, characterized in that, The method for constructing the correspondence between defects and defect conditions includes: Obtain a sample dataset and label the defects in the sample data within the sample dataset; Based on the correspondence between the defects and the feature items, defects and their corresponding feature data are selected from the sample dataset; The selected defects and their corresponding feature data are used as training data to train the random forest classification model, thereby obtaining the correspondence between the defects and the defect conditions.

4. The method according to claim 1, characterized in that, The acquisition of grayscale data of the display panel to be detected includes: Obtain the initial grayscale data of the display panel to be inspected; The initial grayscale data is subjected to threshold processing to obtain the grayscale data.

5. The method according to claim 1, characterized in that, After determining that the display panel has a target defect, the method further includes: Obtain the location information of the target defect from the data that meets the target defect conditions; Based on the location information of the target defect, an image of the target defect is obtained.

6. A defect detection device, characterized in that, include: The first acquisition module is used to acquire grayscale data of the display panel to be detected; The feature extraction module is used to extract features from the grayscale data based on the target defect to be detected, so as to obtain target feature data; The determination module is used to determine the target defect condition corresponding to the target defect from a pre-stored correspondence between defects and defect conditions; The detection module is used to detect whether there is data in the target feature data that meets the target defect conditions; if so, it is determined that the display panel has a target defect. The step of extracting features from the grayscale data based on the target defect to be detected, to obtain target feature data, includes: Based on the pre-stored correspondence between defects and feature terms, the target feature terms corresponding to the target defect are determined; Based on the target feature terms, feature extraction is performed on the grayscale data to obtain target feature data; The method for constructing the correspondence between defects and feature terms includes: Acquire multiple sampling data corresponding to the defect; the sampling data includes multiple parameter items; Among the multiple parameter items, the item that satisfies the correlation condition is taken as the feature item corresponding to the defect, and the correspondence between the defect and the feature item is obtained; the correlation condition includes: in the current parameter item, the proportion of the specified feature data in the item satisfies a first threshold; The step of selecting the item that satisfies the correlation condition from among the plurality of parameter items as the feature item corresponding to the defect includes: When the data in the parameter item is discretely distributed, the parameter item is segmented to obtain multiple candidate data segments; Determine the proportion of the data belonging to the specified defect in each of the candidate data segments; The parameter items corresponding to the candidate data segments whose proportions meet the first threshold are determined as the feature items corresponding to the defect.

7. An electronic device, characterized in that, include: A processor and a memory connected to the processor; The memory is used to store computer programs; The processor is used to call and execute the computer program in the memory to perform the defect detection method as described in any one of claims 1-5.

Images