Panorama-based defect detection method, apparatus and device, and storage medium

Abstract

The invention discloses a defect detection method and device based on a panorama, equipment and a storage medium, and relates to the field of image detection. Obtaining a to-be-detected image set output by defect detection, wherein the to-be-detected image set comprises to-be-reduced-point images recognized and extracted from the panorama of the material plate by other defect detection equipment; selecting a target to-be-reduced-point image from the to-be-detected image set, mapping the target to-be-reduced-point image to the panorama, and extracting a defect standard image corresponding to the to-be-reduced-point image; and respectively identifying and matching contour information in the image to be subjected to point reduction and the defect standard image, and determining defect data and defect information of the image to be subjected to point reduction according to a matching result. According to the scheme, the image to be subjected to point reduction and the mapping position label are selected to be positioned in the position area in the panorama, so that the defect standard image matched with the image to be subjected to point reduction is dug, the defect standard image serves as CAM drawing document data to be verified, missing detection, missing and deviation defects of the contour line are recognized, a model is better optimized, and the product detection rate and accuracy are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image detection, and in particular to a defect detection method, device, equipment and storage medium based on a panoramic image. Background Art

[0002] In the field of semiconductor inspection, there is a kind of inspection process that re-judges the defect images detected by the relevant defect inspection machine to determine whether the original defect image is a "real point" or a "false point", that is, AVS point reduction inspection technology. This inspection usually uses a detection system or inspection machine to verify the defect image output by another detection system or inspection machine, and evaluates the defect detection function of the machine. It is often used for indirect performance comparison of manufacturers and friendly companies' inspection equipment, and re-processes the inspection results to reduce false positives and omissions in the inspection results.

[0003] For defect detection of specific material boards or wafer equipment by competitors, defect detection is usually carried out using CAM drawing data designed one-to-one for the material board to be tested. The CAM drawing data is the CAD master drawing data, which contains the most complete and standard material board outline line drawings. However, during the drop point detection, for protection purposes and other factors, competitors may not provide the corresponding CAM drawing data to the outside. For other manufacturers, they can only use AI recognition and other means to identify the clear defect types in the original drop point image, and cannot accurately identify missed detection situations, especially the omission and deviation of the panoramic image. This also determines that it is difficult to achieve a technical breakthrough in the detection technology. Summary of the invention

[0004] The embodiments of the present application provide a panoramic image-based defect detection method, device, equipment and storage medium to solve the problem of accurately and effectively identifying missing and offset defects in the drop point image without CAM image data.

[0005] On the one hand, the present application provides a defect detection method based on a panoramic image, the method comprising: obtaining a set of images to be inspected output by a defect detection, the set of images to be inspected including images of points to be dropped identified and extracted from a panoramic image of a material plate by other defect detection equipment;

[0006] Selecting a target landing point image from the set of images to be inspected, mapping the target landing point image to the panoramic image, and extracting a defect standard image corresponding to the landing point image;

[0007] The contour information in the image of the waiting point and the defect standard image are identified and matched respectively, and the defect data and defect information of the image of the waiting point are determined according to the matching result.

[0008] Specifically, the image of the waiting landing point extracted from the set of images to be inspected is also embedded with a position tag, and the position tag is used to indicate the coordinate information of the identified image of the waiting landing point in the panoramic image.

[0009] Specifically, mapping the target landing point image to the panoramic image, and extracting the defect standard image corresponding to the landing point image, includes:

[0010] Determine the image size of the target landing point image, and extract the position tag of the target landing point image;

[0011] Mapping the target landing point image into the panoramic image according to the location tag, and defining the image area to be extracted according to the image size;

[0012] The demarcated image area is extracted as the defect standard image.

[0013] Specifically, the location tag records the image center coordinates of the image of the landing point; and mapping the target landing point image into the panoramic image according to the location tag, and defining the image area to be extracted according to the image size, includes:

[0014] Mapping the image center coordinates recorded in the position tag to the panoramic image to determine the center point of the defect standard image;

[0015] Taking the center coordinates of the image as the center, the extracted image area is delineated according to the image size, and the defect standard image is determined and extracted.

[0016] Specifically, the identifying and matching the contour information in the image of the waiting point and the defect standard image respectively, and determining the defect data and defect information of the image of the waiting point according to the matching result, includes:

[0017] Respectively identifying the number of contours contained in the image of the waiting point and the defect standard image; when the number of contours identified in the two images is inconsistent, indicating that there are missed defects in the image of the waiting point;

[0018] When the number of contours identified in the two images is the same, the contours to be inspected in the defect standard image are selected in turn, and contour matching is performed on the defect image to be inspected according to the coordinate information and contour information of the contour to be inspected to determine the defect data and defect information of the image of the point to be inspected.

[0019] Specifically, the contour matching of the defect image to be inspected is performed according to the coordinate information and contour information of the contour to be inspected to determine the defect data and defect information of the image of the point to be dropped, including:

[0020] According to the coordinate information of the contour to be inspected, the contour to be inspected is mapped to the target position in the defect standard map;

[0021] When the contour to be inspected is not matched with the standard image and the target contour at the target position, it is determined that the contour original has a deviation defect.

[0022] Specifically, when the image of the point to be dropped matches the defect standard map and finds a misaligned defect or a missed defect, the corresponding defect information is matched with the defect type output by other defect detection equipment. When the matching results are inconsistent, it is indicated that the current image of the point to be dropped is a false point image.

[0023] On the other hand, the present application provides a defect detection device based on a panoramic image, the device comprising:

[0024] An acquisition module is used to acquire a set of images to be inspected output by defect detection, wherein the set of images to be inspected includes images of points to be dropped that are identified and extracted from a panoramic view of a material board by other defect detection equipment;

[0025] An extraction module, used for selecting a target landing point image from the set of images to be inspected, mapping the target landing point image to the panoramic image, and extracting a defect standard image corresponding to the landing point image;

[0026] The matching module is used to identify and match the contour information in the image of the point to be dropped and the defect standard image respectively, and determine the defect data and defect information of the image of the point to be dropped according to the matching result.

[0027] On the other hand, the present application provides a computer device, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to implement the panoramic image-based defect detection method described in the above aspects.

[0028] On the other hand, the present application provides a computer-readable storage medium, wherein the readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by a processor to implement the panoramic image-based defect detection method described in the above aspects.

[0029] The beneficial effects brought about by the technical solution provided in the embodiments of the present application include at least: making a set of images to be inspected from small defect images output by other AI models or defect detection equipment, and embedding the position label of each image of the waiting point in the set; locating the position area of ​​the target image of the waiting point in the panoramic image by selecting the target image of the waiting point and the mapping position label during re-judgment, thereby extracting a defect standard image that matches it in the panoramic image, and using the defect standard image as CAM drawing data for verification, and determining missed, missing and offset defects by matching the contour lines identified in the two images, thereby better optimizing the model and improving the product detection rate and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of a defect detection method based on a panoramic image provided in an embodiment of the present application;

[0031] Figure 2 It is a schematic diagram of mapping the target waiting point image to the panoramic image;

[0032] Figure 3 A schematic diagram of mapping a panorama based on location tags is shown;

[0033] Figure 4 A structural block diagram of a defect detection device based on a panoramic image is shown;

[0034] Figure 5 A structural block diagram of a computer device provided by an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0035] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0036] The term "multiple" as used herein refers to two or more than two. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the related objects are in an "or" relationship.

[0037] The intelligent vision AVS system is a device independent of the production line. Its purpose is to optimize the performance of the AI ​​defect detection model, specifically by optimizing the images of the waiting point identified by other external defect detection models or external machines. Because the actual wafer fab and the supplier of the AI ​​defect model system are two organizations, large wafer fabs may adopt multiple systems and use them to achieve the effect of mutual promotion between model system suppliers, or in order to make their own AI models achieve better detection results, the model system supplier will compare the target AI model with the defect detection model of a friendly company to make up for its own advantages and disadvantages. At this time, the defect detection model of the friendly company can be run first, and after detecting the wafer material board, a batch of defect images to be confirmed, that is, the images of the waiting point, can be obtained. The images of the waiting point here mean that they may contain normal images, but are mistakenly detected as defective images, or the defect category is incorrectly identified. This application mainly re-judges the situations of missed detection and misalignment of defect contours.

[0038] Figure 1 : is a flow chart of a defect detection method based on a panoramic image provided by an embodiment of the present application, comprising the following steps:

[0039] S1. Obtain a set of images to be inspected output by defect detection, where the set of images to be inspected includes images of points to be dropped that are identified and extracted from a panoramic view of a material board by other defect detection equipment;

[0040] The set of images to be inspected includes a set of several images of waiting points output by friendly companies or other AI defect detection models. This set is the NG images identified by the friendly company's model, and some normal images can also be appropriately added. For the identified NG images or normal images, there may be missed detections and wrong detections due to differences in model progress and performance, especially the missed detection and offset detection defects that are concerned in this application. Missed detection means that the relevant model cannot recognize it, or a certain original component is missing on the material board, and the part that is not accurately identified when the friendly company uses a camera to scan the image and uses the image file data for detection. The offset is due to the insufficient recognition accuracy of the model AI detection. When the contour shape of the original is consistent, but the position is offset, this type of defect cannot be identified. For this solution, the acquired images of the waiting points need to be re-judged one by one to optimize and improve the detection rate.

[0041] In this scheme, based on the AVS output results, the image to be defective is identified again. When the AVS system identifies it as a defective image, it indicates that the image is a true defective image, that is, AVS identifies a true point; when the AVS system identifies it as not a defective image (that is, a normal image), it indicates that the image is a false defective image, that is, AVS identifies a false point.

[0042] In a possible implementation, label information about the image of the waiting point can be embedded in the set of images to be inspected, or a position label can be stored in the set, and the position label is used to indicate the coordinate information of the identified image of the waiting point in the panoramic image. Subsequently, a mapping relationship is established between the label in the set and the image, and a reference re-judgment is performed based on the label information when the system is executed.

[0043] S2, selecting a target landing point image from the set of images to be inspected, mapping the target landing point image to a panoramic image, and extracting a defect standard image corresponding to the landing point image;

[0044] The polling mechanism is used to select the target drop point image in turn, and the corresponding label information is extracted according to the built-in label information or mapping relationship of the image. The panoramic image is mapped according to the label information, that is, to determine which part of the panoramic image this target drop point image corresponds to. Because when detecting defects, the competitors use the CAM image and the panoramic image scanned by their own machine equipment to identify and extract the drop point image from the panoramic image. This solution uses the extracted drop point image to match it in the panoramic image. The panoramic image provided in this solution should be a standard and correct panoramic image produced by the wafer manufacturer or by a high-precision camera scan before implementing this solution. Special attention should be paid to the fact that the panoramic image here is different from the CAM image, because the CAM image is a master image designed one-to-one before the production of the wafer or material, that is, the content is a simple vector contour line with high precision, and the panoramic image is a photo image obtained by scanning the wafer or material board. Therefore, if additional defect information can be identified on the basis of the panoramic image, it is sufficient to prove the superiority of this method.

[0045] Figure 2 It is a schematic diagram of mapping the target landing point image to the panoramic image. When the product model and structure are consistent, the selected target landing point image can be mapped into the panoramic image. The mapped area is equivalent to the standard small image, that is, the defect standard image corresponding to the landing point image is extracted. This defect standard image is a standard image that theoretically does not have any defects. It serves as CAM image file data and is used for later matching and identification.

[0046] S3, respectively identifying and matching the contour information in the image of the point to be dropped and the defect standard image, and determining the defect data and defect information of the image of the point to be dropped according to the matching results.

[0047] The matching method mainly maps the contour lines of the original in the defect standard image to the image of the point to be dropped, or maps the contour lines of the original in the image of the point to be dropped to the defect standard image, and determines the defect data and defect information through the difference changes in the line contours of the two. For the original that is missing in the image of the point to be dropped, it is impossible to match one to one during matching. When the shape, length and position of the two contour lines are inconsistent, it can be determined that the image of the point to be dropped has missing, deformed and offset defects. These defect data and defect information can be stored and output in the form of labels for subsequent manual confirmation.

[0048] To summarize, the present application uses the defect images output by other AI models or defect detection equipment to create a set of images to be inspected, and embeds the position label of each image of the landing point to be inspected in the set; during the re-judgment, the target landing point image and the mapped position label are selected to locate the position area of ​​the target landing point image in the panoramic image, so as to extract the defect standard image that matches it in the panoramic image, and use the defect standard image as CAM drawing data for verification, and determine the missed, missing and offset defects by matching the contour lines identified in the two images, so as to better optimize the model and improve the product detection rate and accuracy.

[0049] In some embodiments, the process of mapping the target landing point image to the panoramic image and extracting the defect standard image corresponding to the landing point image may specifically include the following steps:

[0050] S21, determining the image size of the target landing point image, and extracting the position label of the target landing point image;

[0051] S22, mapping the target landing point image into the panoramic image according to the position tag, and defining the image area to be extracted according to the image size;

[0052] This step is to extract a defect standard image that is consistent with the size of the target drop point image. Because it involves pixel matching, extraction and mapping must meet high precision requirements. For this reason, this application chooses to use the image center coordinates of the extracted drop point image to make a position label and embed the label into the drop point image. Figure 3 The diagram shows a schematic diagram of mapping a panoramic image based on a position label. The image on the left represents the image center of the extracted target drop point image, and the coordinates of the defect point are not used as the position label. Because the identified defects are of different sizes and shapes, and the defect point is not necessarily a true defect, it is more accurate to use the image center mapping. First, the image center coordinates recorded in the position label are mapped to the panoramic image to determine the center point of the defect standard image. Then, the extracted image area is delineated according to the image size based on the image center coordinates and determined as the defect standard image.

[0053] S23, extracting the designated image area as a defect standard image.

[0054] In some other embodiments, the process of identifying and matching the contour information in the image of the point to be dropped and the defect standard image, and determining the defect data and defect information of the image of the point to be dropped according to the matching result can be specifically summarized as follows:

[0055] S31, respectively identifying the number of contours contained in the image of the waiting point and the defect standard image; when the number of contours identified in the two images is inconsistent, indicating that there is a missed defect in the image of the waiting point;

[0056] Contour recognition can be extracted and confirmed using AI recognition algorithms. After extracting the contours, it is not directly matched, but the number of contours in the two images is determined first. When the number of contours in the two images is inconsistent, there must be a situation where the original is missing or missed. Especially for the situation where the original output is defect-free, this step directly identifies the existence of missed defects, indicating that the original model has limited accuracy.

[0057] S32. When the number of contours identified in the two images is the same, the contours to be inspected in the defect standard image are selected in turn, and contour matching is performed on the defect image to be inspected according to the coordinate information and contour information of the contour to be inspected, so as to determine the defect data and defect information of the image of the point to be inspected.

[0058] When the number of contours identified in the two images is the same, it indicates that there is no missed detection. It is necessary to further determine whether there are other defects such as deviation. At this time, it is necessary to select the contours to be detected in the defect standard image in turn. This step requires obtaining the coordinate information of each identified contour in advance, such as determining the coordinates with the starting segment of the contour line, and then mapping one of the contour lines to the other image according to the coordinates for alignment matching. From the pixel level analysis, when the positions and shapes of the two contour lines are exactly the same, the coordinates and contour lines are completely overlapped, or the deviation is within a certain threshold range. When the offset between the two contour lines is large and the coordinate positions are inconsistent, it indicates that there is wear and tear, as well as deviation. This defect may not be detected in the original AI model. When the image of the point to be dropped matches the defect standard image for a deviation defect or a missed defect, the corresponding defect information is matched with the defect type output by other defect detection equipment. When the matching results are inconsistent, especially when the image of the point to be dropped is normal and no problem is detected, but a deviation is identified after matching, it means that the current image of the point to be dropped is a false point image.

[0059] Figure 4 The structural block diagram of a panoramic image-based defect detection device provided in an embodiment of the present application is shown, and the device includes:

[0060] An acquisition module 410 is used to acquire a set of images to be inspected output by defect detection, wherein the set of images to be inspected includes images of points to be dropped that are identified and extracted from a panoramic view of a material plate by other defect detection equipment;

[0061] An extraction module 420 is used to select a target landing point image from the set of images to be inspected, map the target landing point image to the panoramic image, and extract a defect standard image corresponding to the landing point image;

[0062] The matching module 430 is used to identify and match the contour information in the image of the waiting point and the defect standard image respectively, and determine the defect data and defect information of the image of the waiting point according to the matching result.

[0063] The panoramic image-based defect detection device provided in the embodiment of the present application can be applied to the panoramic image-based defect detection method provided in the above-mentioned embodiment. For relevant details, refer to the above-mentioned method embodiment. The implementation principle and technical effect are similar and will not be repeated here.

[0064] It should be noted that the panoramic image-based defect detection device provided in the embodiments of the present application is only illustrated by the division of the above-mentioned functional modules / functional units. In actual applications, the above-mentioned functions can be assigned to different functional modules / functional units as needed, that is, the internal structure of the panoramic image-based defect detection device is divided into different functional modules / functional units to complete all or part of the functions described above. In addition, the implementation method of the panoramic image-based defect detection method provided in the above-mentioned method embodiment and the implementation method of the panoramic image-based defect detection device provided in this embodiment belong to the same concept. The specific implementation process of the panoramic image-based defect detection device provided in this embodiment is detailed in the above-mentioned method embodiment, which will not be repeated here.

[0065] Figure 5The block diagram of a computer device provided by an exemplary embodiment of the present application is shown. It is a computer device such as a desktop computer, a laptop computer, a PDA, and a cloud server. The computer device may include, but is not limited to, a processor and a memory. Among them, the processor and the memory may be connected by a bus or in other ways. Among them, the processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, graphics processors (GPU), embedded neural network processors (NPU) or other dedicated deep learning coprocessors, discrete gates or transistor logic devices, discrete hardware components and other chips, or a combination of the above-mentioned various chips.

[0066] The processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content that needs to be displayed on the display screen. In some embodiments, the processor may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0067] As a non-transient computer-readable storage medium, the memory can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the methods in the above-mentioned embodiments of the present application. The processor executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory, that is, the method in the above-mentioned method implementation is realized. The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0068] In some embodiments, the computer device may further optionally include: a peripheral device interface and at least one peripheral device. The processor, the memory and the peripheral device interface may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit, a display screen and a keyboard.

[0069] The peripheral device interface can be used to connect at least one peripheral device related to I / O (Input / Output) to the processor and the memory. In some embodiments, the processor, the memory, and the peripheral device interface are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor, the memory, and the peripheral device interface can be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0070] The display screen is used to display the UI (User Interface). The UI may include graphics, text, icons, videos and any combination thereof. When the display screen is a touch screen, the display screen also has the ability to collect touch signals on the surface or above the surface of the display screen. The touch signal can be input to the processor as a control signal for processing. At this time, the display screen can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen can be one, arranged on the front panel of the computer device; in other embodiments, the display screen can be at least two, respectively arranged on different surfaces of the computer device or in a folding design; in other embodiments, the display screen can be a flexible display screen, arranged on a curved surface or a folding surface of the computer device. Even, the display screen can be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode) and the like.

[0071] The power supply is used to power various components in the computer device. The power supply can be AC, DC, a disposable battery, or a rechargeable battery. When the power supply includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged through a wired line, and a wireless rechargeable battery is a battery that is charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0072] Those skilled in the art will appreciate that the structure shown in this embodiment does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or adopt a different component arrangement.

[0073] The embodiment of the present application also discloses a computer-readable storage medium. Specifically, the computer-readable storage medium is used to store a computer program, and when the computer program is executed by the processor, the method in the above-mentioned method implementation is implemented. Those skilled in the art can understand that the implementation of all or part of the process in the above-mentioned implementation method of the present application can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the process of the implementation of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviated as: HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0074] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A defect detection method based on a panoramic image, characterized in that: The method comprises: Acquire a set of images to be inspected output by defect detection, wherein the set of images to be inspected includes images of points to be dropped that are identified and extracted from a panoramic view of a material plate by other defect detection equipment; Selecting a target landing point image from the set of images to be inspected, mapping the target landing point image to the panoramic image, and extracting a defect standard image corresponding to the landing point image; The contour information in the image of the waiting point and the defect standard image are identified and matched respectively, and the defect data and defect information of the image of the waiting point are determined according to the matching result.

2. The method according to claim 1, characterized in that The image of the waiting landing point extracted from the set of images to be inspected is also embedded with a position tag, and the position tag is used to indicate the coordinate information of the identified image of the waiting landing point in the panoramic view.

3. The method according to claim 2, characterized in that The step of mapping the target landing point image to the panoramic image and extracting the defect standard image corresponding to the landing point image includes: Determine the image size of the target landing point image, and extract the position tag of the target landing point image; Mapping the target landing point image into the panoramic image according to the location tag, and defining the image area to be extracted according to the image size; The demarcated image area is extracted as the defect standard image.

4. The method according to claim 3, characterized in that The location tag records the image center coordinates of the image of the landing point; the mapping of the target landing point image into the panoramic image according to the location tag, and defining the image area to be extracted according to the image size, includes: Mapping the image center coordinates recorded in the position tag to the panoramic image to determine the center point of the defect standard image; Taking the center coordinates of the image as the center, the extracted image area is delineated according to the image size, and the defect standard image is determined and extracted.

5. The method according to claim 3, characterized in that: The step of respectively identifying and matching the contour information in the image of the waiting point and the defect standard image, and determining the defect data and defect information of the image of the waiting point according to the matching result, includes: Respectively identifying the number of contours contained in the image of the waiting point and the defect standard image; when the number of contours identified in the two images is inconsistent, indicating that there are missed defects in the image of the waiting point; When the number of contours identified in the two images is the same, the contours to be inspected in the defect standard image are selected in turn, and contour matching is performed on the defect image to be inspected according to the coordinate information and contour information of the contour to be inspected to determine the defect data and defect information of the image of the point to be inspected.

6. The method according to claim 5, characterized in that The step of performing contour matching on the defect image to be inspected according to the coordinate information and contour information of the contour to be inspected to determine the defect data and defect information of the image of the point to be inspected includes: According to the coordinate information of the contour to be inspected, the contour to be inspected is mapped to the target position in the defect standard map; When the contour to be inspected is not matched with the standard image and the target contour at the target position, it is determined that the contour original has a deviation defect.

7. The method according to any one of claims 1 to 6, characterized in that: When the image of the waiting point matches the defect standard map and finds a misaligned defect or a missed defect, the corresponding defect information is matched with the defect type output by other defect detection equipment. When the matching results are inconsistent, it is indicated that the current image of the waiting point is a false point image.

8. A defect detection device based on a panoramic image, characterized in that: The device comprises: An acquisition module is used to acquire a set of images to be inspected output by defect detection, wherein the set of images to be inspected includes images of points to be dropped that are identified and extracted from a panoramic view of a material board by other defect detection equipment; An extraction module, used for selecting a target landing point image from the set of images to be inspected, mapping the target landing point image to the panoramic image, and extracting a defect standard image corresponding to the landing point image; The matching module is used to identify and match the contour information in the image of the point to be dropped and the defect standard image respectively, and determine the defect data and defect information of the image of the point to be dropped according to the matching result.

9. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the panoramic image-based defect detection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to implement the panoramic image-based defect detection method as described in any one of claims 1 to 7.

Images