Method and system for identifying polyimide residual defects in chip packaging process

Abstract

The invention provides a method and a system for identifying polyimide residual defects in a chip packaging process, and relates to the technical field of defect detection, and the method comprises the following steps: carrying out pixel splicing on a chip color image and a chip fluorescence image to obtain a color fluorescence spliced image; inputting the chip color image into a deep learning model for foreign matter defect positioning to obtain foreign matter defect positioning information; carrying out pixel value verification on the chip fluorescence image to obtain polyimide residual defect positioning information; mapping the foreign matter defect positioning information and the polyimide residual defect positioning information to the color fluorescent spliced image to obtain a defect spliced mapping image; and performing logic verification on the defect splicing mapping image to obtain a polyimide residual defect identification result. According to the method, automatic detection of polyimide residual defects can be achieved, the detection efficiency is high, defect detection is carried out in a parameter comparison and verification mode, and therefore the problems of missing detection and false detection cannot occur.

Description

Technical Field

[0001] The present invention relates to the field of defect detection technology, and in particular to a method and system for identifying polyimide residue defects in a chip packaging process. Background Art

[0002] Chips, also known as IC (Integrated Circuit), are a high-tech industry that is being vigorously developed and researched by all countries in the world today. The chip industry mainly includes three parts: chip design, chip manufacturing, and chip packaging and testing. Detection and control of defects generated in different process steps during chip packaging is an important part of improving chip packaging technology in chip packaging and testing.

[0003] The rules for judging polyimide residue defects generated at the same process site during chip packaging are relatively complex. Manual image judgment requires attention to both color images and fluorescent images. The cost of manual image judgment is high and is affected by subjective factors of the judges, resulting in low image judgment efficiency and prone to missed detections and false detections. Summary of the invention

[0004] In order to solve the problem that the recognition efficiency of polyimide residue defects in a chip packaging process is low and missed detection and false detection are prone to occur, an embodiment of the present invention provides a method and system for recognizing polyimide residue defects in a chip packaging process.

[0005] In a first aspect, an embodiment of the present invention provides a method for identifying polyimide residue defects in a chip packaging process, and the method flow is as follows:

[0006] Acquire a chip color image and a chip fluorescence image of the same process station during the chip packaging process, and perform pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitching image;

[0007] The chip color image is input into the deep learning model to locate foreign matter defects to obtain foreign matter defect location information; and the chip fluorescence image is verified by pixel value to obtain polyimide residue defect location information;

[0008] Mapping foreign matter defect location information and polyimide residue defect location information to the color fluorescent stitching image to obtain a defect stitching mapping image;

[0009] The defect splicing mapping image is logically checked according to the polyimide residue defect verification rules to obtain the final polyimide residue defect recognition result.

[0010] In the above embodiment, the present invention first splices the chip color image and the chip fluorescence image to obtain a color fluorescence spliced ​​image; then, defect positioning is performed on the chip color image and the chip fluorescence image respectively to obtain foreign matter defect positioning information and polyimide residue defect positioning information, and then verification parameter mapping is performed according to the foreign matter defect positioning information and the polyimide residue defect positioning information to obtain a defect splicing mapping image, and finally, a logical check is performed on the defect splicing mapping image to obtain different defect identification results, which can realize automatic detection of polyimide residue defects, has high inspection efficiency, and adopts parameter comparison and verification to perform defect detection, so there will be no problems of missed detection and false detection.

[0011] As some optional implementations of the present application, the process of performing pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitched image is as follows:

[0012] Dividing the chip color image and the chip fluorescence image according to single pixels or pixel blocks to obtain a plurality of chip color sub-images and chip fluorescence sub-images;

[0013] The chip color sub-image and the chip fluorescence sub-image are pixel-stitched according to the image division order to obtain a color fluorescence stitching image.

[0014] In the above embodiment, the present invention fuses the chip color image and the chip fluorescence image in a division and splicing manner, so that the subsequent parameter comparison and verification is more convenient and the problems of false detection and missed detection will not occur.

[0015] As some optional implementations of the present application, before inputting the chip color image into the deep learning model for foreign body defect location, the deep learning model needs to be trained.

[0016] In the above embodiment, the present invention trains the deep learning model so that the deep learning model can automatically identify foreign matter defects in the color image of the chip.

[0017] As some optional implementation methods of the present application, the process of training the deep learning model is as follows:

[0018] Acquire a number of chip color images, and label the chip color images with foreign matter defect categories to obtain a training sample set;

[0019] The chip color images labeled with foreign body defect categories in the training sample set are input into the deep learning model for model training to obtain the global pooling features of the foreign body defect categories.

[0020] As some optional implementations of the present application, the foreign body defect positioning information includes coordinate information of all foreign body defects.

[0021] In the above embodiment, the present invention can facilitate the subsequent mapping of verification parameters by extracting the coordinate information of the foreign matter defect.

[0022] As some optional implementations of the present application, before performing pixel value verification on the chip fluorescence image, it is necessary to perform image grayscale conversion on the chip fluorescence image.

[0023] In the above embodiment, the present invention makes the defect contour of the chip fluorescence image more obvious through image grayscale conversion.

[0024] As some optional implementations of the present application, the process of performing pixel value verification on the chip fluorescence image to obtain the location information of polyimide residue defects is as follows:

[0025] Perform pixel threshold check on the pixel points of the chip fluorescence image. If the pixel value of the pixel point is greater than the pixel threshold, the coordinate information of the pixel point is retained; otherwise, the coordinate information of the pixel point is removed.

[0026] All pixel points of the chip fluorescence image are traversed to obtain polyimide residue defect location information, wherein the polyimide residue defect location information includes coordinate information of all polyimide residue defects.

[0027] In the above embodiment, the present invention can quickly and accurately locate defects by performing pixel threshold verification on pixel points of the chip fluorescence image.

[0028] As some optional implementations of the present application, the process of mapping the foreign matter defect location information and the polyimide residue defect location information to the color fluorescent spliced ​​image to obtain the defect spliced ​​mapping image is as follows:

[0029] Setting a first verification parameter for a pixel point corresponding to the chip color image according to the foreign body defect positioning information to obtain a chip color image including the first verification parameter;

[0030] According to the polyimide residue defect location information, a second verification parameter is set for the pixel points corresponding to the chip fluorescence image to obtain a chip fluorescence image including the second verification parameter;

[0031] According to the pixel stitching order of the chip color image and the chip fluorescence image, the first verification parameter of the chip color image and the second verification parameter of the chip fluorescence image are mapped to the color fluorescence stitching image to obtain a defect stitching mapping image including the first verification parameter and the second verification parameter.

[0032] In the above embodiment, the present invention assigns verification parameters to the chip color image and the chip fluorescence image according to the foreign matter defect location information and the polyimide residue defect location information, which can facilitate the subsequent mapping of defect information.

[0033] As some optional implementations of the present application, the process of performing logic verification processing on the defect splicing mapping image according to the polyimide residue defect verification rules to obtain the final polyimide residue defect recognition result is as follows:

[0034] Performing first verification parameter and second verification parameter detection on corresponding pixel points of the defective splicing mapping image;

[0035] If the first verification parameter and the second verification parameter exist at the corresponding pixel point of the color fluorescent stitching image at the same time, it is determined that there is a polyimide residue defect caused by foreign matter;

[0036] If the first verification parameter exists but the second verification parameter does not exist at the pixel point corresponding to the color fluorescent stitching image, it is determined that a foreign body defect exists;

[0037] If the first verification parameter does not exist for the pixel point corresponding to the color fluorescent stitching image but the second verification parameter exists, it is determined that there is a polyimide residue defect not caused by foreign matter.

[0038] In the above embodiment, the present invention performs logic verification on the first verification parameter and the second verification parameter to obtain different defect identification results.

[0039] In a second aspect, the present invention provides a polyimide residue defect identification system in a chip packaging process, wherein the functions of the system correspond to the method one by one, and the system comprises:

[0040] A defect image stitching unit, the defect image stitching unit is used to obtain a chip color image and a chip fluorescence image of the same process station in the chip packaging process, and perform pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitching image;

[0041] A defect location information acquisition unit, wherein the defect location information acquisition unit is used to input the chip color image into the deep learning model to locate foreign body defects to obtain foreign body defect location information; and to perform pixel value verification on the chip fluorescence image to obtain polyimide residue defect location information;

[0042] A defect location information mapping unit, wherein the defect location information mapping unit is used to map foreign matter defect location information and polyimide residue defect location information to a color fluorescent spliced ​​image to obtain a defect spliced ​​mapping image;

[0043] The defect category recognition unit is used to perform a logic check on the defect splicing mapping image according to the polyimide residue defect verification rule to obtain a final polyimide residue defect recognition result.

[0044] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for identifying polyimide residue defects in the chip packaging process.

[0045] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, a method for identifying polyimide residue defects in the chip packaging process is implemented.

[0046] The beneficial effects of the present invention are as follows: the present invention first splices the chip color image and the chip fluorescence image to obtain a color fluorescence spliced ​​image; then performs defect location on the chip color image and the chip fluorescence image respectively to obtain foreign body defect location information and polyimide residue defect location information, and then performs verification parameter mapping according to the foreign body defect location information and the polyimide residue defect location information to obtain a defect splicing mapping image, and finally performs logical verification on the defect splicing mapping image to obtain different defect recognition results, which can realize automatic detection of polyimide residue defects, has high inspection efficiency, and adopts parameter comparison verification to perform defect detection, so there will be no problems of missed detection and false detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 is a schematic diagram of the computer device structure of the hardware operating environment described in the embodiment of the present invention;

[0049] Figure 2 is a flow chart of a method for identifying polyimide residue defects in a chip packaging process according to an embodiment of the present invention;

[0050] Figure 3 is a schematic diagram of pixel stitching of a chip color image and a chip fluorescence image according to an embodiment of the present invention;

[0051] Figure 4 It is a structural block diagram of a polyimide residue defect recognition system in a chip packaging process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] In order to solve the problem of low efficiency in identifying polyimide residue defects in the chip packaging process, which is prone to missed detection and false detection, the present application provides a method and system for identifying polyimide residue defects in the chip packaging process. Before introducing the specific technical solution of the present application, the hardware operating environment involved in the embodiment of the present application is first introduced.

[0054] See also Figure 1 , Figure 1 A schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0055] like Figure 1 As shown, the computer device may include: a processor, such as a central processing unit (CPU), a communication bus, a user interface, a network interface, and a memory. Among them, the communication bus is used to realize the connection and communication between these components. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), and optionally, the user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity interface). The memory may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory may optionally be a storage device independent of the aforementioned processor.

[0056] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0057] like Figure 1 As shown, the memory as a storage medium may include an operating system module, a network communication module, a user interface module and a data storage module.

[0058] exist Figure 1 In the computer device shown, the network interface is mainly used for data communication with the network server; the user interface is mainly used for data interaction with the user; the processor and memory in the computer device of the present application can be set in the computer device, and the computer device calls the polyimide residue defect identification system in the chip packaging process stored in the data storage module in the memory through the processor, and executes the polyimide residue defect identification method in the chip packaging process provided in the embodiment of the present application.

[0059] Based on the hardware environment of the above-mentioned embodiment, the embodiment of the present application provides a method for identifying polyimide residue defects in a chip packaging process. Figure 2 , Figure 2 The flowchart of the method for identifying polyimide residue defects in the chip packaging process is as follows:

[0060] (1) Obtain a chip color image and a chip fluorescence image of the same process station during the chip packaging process, and perform pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitching image.

[0061] In the embodiment of the present invention, please refer to Figure 3 , Figure 3 Schematic diagram of pixel stitching of chip color image and chip fluorescence image, where: Figure 3 (a) represents a color image. Figure 3 (b) shows the fluorescence image of the chip. Figure 3 (c) represents a color fluorescence stitching image. Specifically, the process of performing pixel stitching on the chip color image and the chip fluorescence image is as follows:

[0062] (1.1) The chip color image and the chip fluorescence image are divided according to single pixels or pixel blocks to obtain a plurality of chip color sub-images and chip fluorescence sub-images; that is, the chip color image and the chip fluorescence image are divided horizontally or vertically into a plurality of chip color sub-images and chip fluorescence sub-images of exactly the same shape and size according to a certain pixel width.

[0063] (1.2) The chip color sub-image and the chip fluorescence sub-image are pixel-stitched in the order of image division to obtain a color fluorescence stitched image; that is, the chip color sub-image and the chip fluorescence sub-image are stitched from left to right or from top to bottom in sequence in the order of image division, so that the sub-images at the same position can be arranged side by side left to right or side by side up and down to obtain a color fluorescence stitched image containing relevant information of the chip color sub-image and the chip fluorescence sub-image.

[0064] (2) Inputting the chip color image into the deep learning model to locate foreign matter defects to obtain foreign matter defect location information; performing image grayscale conversion processing on the chip fluorescence image, and performing pixel value verification on the chip fluorescence image to obtain polyimide residue defect location information, wherein the foreign matter defect location information includes the coordinate information of all foreign matter defects, and the polyimide residue defect location information includes the coordinate information of all polyimide residue defects.

[0065] In an embodiment of the present invention, before inputting the chip color image into the deep learning model to locate foreign body defects, the deep learning model needs to be trained.

[0066] Specifically, the process of model training for the deep learning model is as follows:

[0067] (2.1) Obtain a number of chip color images and label the chip color images with foreign body defect categories to obtain a training sample set.

[0068] (2.2) Inputting chip color images labeled with foreign body defect categories in the training sample set into a deep learning model for model training to obtain global pooling features of the foreign body defect categories, wherein the deep learning model includes but is not limited to a YOLOV5 target detection model.

[0069] Specifically, the process of verifying the pixel value of the chip fluorescence image is as follows:

[0070] (2.3) Performing pixel threshold check on the pixel points of the chip fluorescence image. If the pixel value of the pixel point is greater than the pixel threshold, the coordinate information of the pixel point is retained. Otherwise, the coordinate information of the pixel point is removed. That is, if the pixel value of the pixel point is greater than the pixel threshold, it can be determined that there is a defect and the coordinate information of the corresponding pixel point is retained. Otherwise, it can be determined that there is no defect and there is no need to retain the coordinate information of the corresponding pixel point. The pixel threshold can be set according to actual conditions, including but not limited to any value between 200 and 255.

[0071] (2.4) Traversing all the pixels of the chip fluorescence image, the location information of polyimide residue defects can be obtained by performing pixel threshold verification on all the pixels.

[0072] (3) Mapping the foreign matter defect location information and the polyimide residue defect location information to the color fluorescent stitching image to obtain a defect stitching mapping image.

[0073] In an embodiment of the present invention, the process of mapping the foreign matter defect location information and the polyimide residue defect location information to the color fluorescent stitching image is as follows:

[0074] (3.1) Setting a first verification parameter for the corresponding pixel points of the chip color image according to the foreign matter defect positioning information to obtain a chip color image containing the first verification parameter; that is, according to the foreign matter defect positioning information, the first verification parameter of the corresponding pixel point can be set to 1, and the second verification parameter of the remaining pixel points can be set to 0.

[0075] (3.2) According to the polyimide residue defect positioning information, the second verification parameter of the corresponding pixel point of the chip fluorescence image is set to obtain a chip fluorescence image containing the second verification parameter; that is, according to the polyimide residue defect positioning information, the second verification parameter of the corresponding pixel point can be set to 1, and the second verification parameters of the remaining pixel points can be set to 0.

[0076] (3.3) Mapping the first verification parameter of the chip color image and the second verification parameter of the chip fluorescence image to the color fluorescence stitching image according to the pixel stitching order of the chip color image and the chip fluorescence image to obtain a defect stitching mapping image including the first verification parameter and the second verification parameter.

[0077] (4) Performing a logical check on the defect splicing mapping image according to the polyimide residue defect verification rules to obtain the final polyimide residue defect recognition result.

[0078] In an embodiment of the present invention, the process of performing logic verification on the defective splicing mapping image according to the polyimide residue defect verification rule is as follows:

[0079] (4.1) Performing a first verification parameter and a second verification parameter detection on the corresponding pixel points of the defective splicing mapping image.

[0080] (4.2) If the first verification parameter and the second verification parameter exist at the same time for the corresponding pixel point of the color fluorescent stitching image, it is determined that there is a polyimide residue defect caused by foreign matter; that is, if the first verification parameter and the second verification parameter of the pixel point at the same position are both 1, it is determined that there is a polyimide residue defect caused by foreign matter at the corresponding position.

[0081] (4.3) If the first verification parameter exists for the corresponding pixel point of the color fluorescent stitching image but the second verification parameter does not exist, it is determined that a foreign body defect exists. That is, if the first verification parameter of the pixel point at the same position is 1 and the second verification parameter is 0, it is determined that a foreign body defect exists at the corresponding position.

[0082] (4.4) If the first verification parameter does not exist for the corresponding pixel point of the color fluorescent stitching image but the second verification parameter exists, it is determined that there is a polyimide residue defect not caused by foreign matter, that is, if the first verification parameter of the pixel point at the same position is 0 and the second verification parameter is 1, it is determined that there is a polyimide residue defect not caused by foreign matter at the corresponding position.

[0083] In summary, the present invention first splices the chip color image and the chip fluorescence image to obtain a color fluorescence spliced ​​image; then performs defect location on the chip color image and the chip fluorescence image respectively to obtain foreign matter defect location information and polyimide residue defect location information, and then performs verification parameter mapping according to the foreign matter defect location information and the polyimide residue defect location information to obtain a defect splicing mapping image, and finally performs logical verification on the defect splicing mapping image to obtain different defect recognition results, which can realize automatic detection of polyimide residue defects, has high inspection efficiency, and adopts parameter comparison verification to perform defect detection, so there will be no problems of missed detection and false detection.

[0084] In addition, in one embodiment, based on the same inventive concept as the above embodiment, the embodiment of the present invention provides a polyimide residue defect recognition system in a chip packaging process, and the system corresponds to the method of embodiment 1 one by one, please refer to Figure 4 , Figure 4 The structure block diagram of the polyimide residue defect recognition system in the chip packaging process is as follows:

[0085] A defect image stitching unit, the defect image stitching unit is used to obtain a chip color image and a chip fluorescence image of the same process station in the chip packaging process, and perform pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitching image;

[0086] A defect location information acquisition unit, wherein the defect location information acquisition unit is used to input the chip color image into the deep learning model to locate foreign body defects to obtain foreign body defect location information; and to perform pixel value verification on the chip fluorescence image to obtain polyimide residue defect location information;

[0087] A defect location information mapping unit, wherein the defect location information mapping unit is used to map foreign matter defect location information and polyimide residue defect location information to a color fluorescent spliced ​​image to obtain a defect spliced ​​mapping image;

[0088] The defect category recognition unit is used to perform a logic check on the defect splicing mapping image according to the polyimide residue defect verification rule to obtain a final polyimide residue defect recognition result.

[0089] It should be noted that each unit in the polyimide residue defect identification system during chip packaging in this embodiment corresponds one-to-one to each step in the polyimide residue defect identification method during chip packaging in the aforementioned embodiment. Therefore, the specific implementation method and technical effect achieved by this embodiment can refer to the implementation method of the polyimide residue defect identification method during chip packaging in the aforementioned chip packaging process, and will not be repeated here.

[0090] In addition, in one embodiment, the present application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory, and the computer program implements the method in the aforementioned embodiment when executed by the processor.

[0091] In addition, in one embodiment, the present application further provides a computer storage medium, on which a computer program is stored, and the computer program implements the method in the aforementioned embodiment when executed by a processor.

[0092] In some embodiments, the computer readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or various devices including one or any combination of the above memories. The computer may be various computing devices including intelligent terminals and servers.

[0093] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0094] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0095] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0096] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0097] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0098] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0099] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for identifying polyimide residue defects in a chip packaging process, characterized in that: The method flow is as follows: Acquire a chip color image and a chip fluorescence image of the same process station during the chip packaging process, and perform pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitching image; The chip color image is input into the deep learning model to locate foreign matter defects to obtain foreign matter defect location information; and the chip fluorescence image is verified by pixel value to obtain polyimide residue defect location information; Mapping foreign matter defect location information and polyimide residue defect location information to the color fluorescent stitching image to obtain a defect stitching mapping image; The defect splicing mapping image is logically checked according to the polyimide residue defect verification rules to obtain the final polyimide residue defect recognition result.

2. The method for identifying polyimide residue defects in a chip packaging process according to claim 1, characterized in that: The process of pixel stitching of the chip color image and the chip fluorescence image to obtain the color fluorescence stitching image is as follows: Dividing the chip color image and the chip fluorescence image according to single pixels or pixel blocks to obtain a plurality of chip color sub-images and chip fluorescence sub-images; The chip color sub-image and the chip fluorescence sub-image are pixel-stitched according to the image division order to obtain a color fluorescence stitching image.

3. The method for identifying polyimide residue defects in a chip packaging process according to claim 1, characterized in that: Before inputting the chip color image into the deep learning model to locate foreign body defects, the deep learning model needs to be trained.

4. The method for identifying polyimide residue defects in a chip packaging process according to claim 3, characterized in that: The process of model training for a deep learning model is as follows: Acquire a number of chip color images, and label the chip color images with foreign matter defect categories to obtain a training sample set; The chip color images labeled with foreign body defect categories in the training sample set are input into the deep learning model for model training to obtain the global pooling features of the foreign body defect categories.

5. The method for identifying polyimide residue defects in a chip packaging process according to claim 1, characterized in that: The foreign body defect positioning information includes coordinate information of all foreign body defects.

6. The method for identifying polyimide residue defects in a chip packaging process according to claim 1, characterized in that: Before performing pixel value verification on the chip fluorescence image, the chip fluorescence image needs to be converted into grayscale.

7. The method for identifying polyimide residue defects in a chip packaging process according to claim 1, characterized in that: The process of verifying the pixel values ​​of the chip fluorescence image to obtain the location information of polyimide residue defects is as follows: Perform pixel threshold check on the pixel points of the chip fluorescence image. If the pixel value of the pixel point is greater than the pixel threshold, the coordinate information of the pixel point is retained; otherwise, the coordinate information of the pixel point is removed. All pixel points of the chip fluorescence image are traversed to obtain polyimide residue defect location information, wherein the polyimide residue defect location information includes coordinate information of all polyimide residue defects.

8. The method for identifying polyimide residue defects in a chip packaging process according to claim 1, characterized in that: The process of mapping the foreign matter defect location information and the polyimide residue defect location information to the color fluorescent stitching image to obtain the defect stitching mapping image is as follows: Setting a first verification parameter for a pixel point corresponding to the chip color image according to the foreign body defect positioning information to obtain a chip color image including the first verification parameter; According to the polyimide residue defect location information, a second verification parameter is set for the pixel points corresponding to the chip fluorescence image to obtain a chip fluorescence image including the second verification parameter; According to the pixel stitching order of the chip color image and the chip fluorescence image, the first verification parameter of the chip color image and the second verification parameter of the chip fluorescence image are mapped to the color fluorescence stitching image to obtain a defect stitching mapping image including the first verification parameter and the second verification parameter.

9. The method for identifying polyimide residue defects in a chip packaging process according to claim 8, characterized in that: The process of performing logical verification on the defect splicing mapping image according to the polyimide residue defect verification rules to obtain the final polyimide residue defect recognition result is as follows: Performing first verification parameter and second verification parameter detection on corresponding pixel points of the defective splicing mapping image; If the first verification parameter and the second verification parameter exist at the corresponding pixel point of the color fluorescent stitching image at the same time, it is determined that there is a polyimide residue defect caused by foreign matter; If the first verification parameter exists but the second verification parameter does not exist at the pixel point corresponding to the color fluorescent stitching image, it is determined that a foreign body defect exists; If the first verification parameter does not exist for the pixel point corresponding to the color fluorescent stitching image but the second verification parameter exists, it is determined that there is a polyimide residue defect not caused by foreign matter.

10. A polyimide residue defect recognition system in a chip packaging process, characterized in that: The system comprises: A defect image stitching unit, the defect image stitching unit is used to obtain a chip color image and a chip fluorescence image of the same process station in the chip packaging process, and perform pixel stitching on the chip color image and the chip fluorescence image to obtain a color fluorescence stitching image; A defect location information acquisition unit, wherein the defect location information acquisition unit is used to input the chip color image into the deep learning model to locate foreign body defects to obtain foreign body defect location information; and to perform pixel value verification on the chip fluorescence image to obtain polyimide residue defect location information; A defect location information mapping unit, wherein the defect location information mapping unit is used to map foreign matter defect location information and polyimide residue defect location information to a color fluorescent spliced ​​image to obtain a defect spliced ​​mapping image; The defect category recognition unit is used to perform a logic check on the defect splicing mapping image according to the polyimide residue defect verification rule to obtain a final polyimide residue defect recognition result.

11. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for identifying polyimide residue defects in a chip packaging process as described in any one of claims 1 to 9 is implemented.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for identifying polyimide residue defects in a chip packaging process according to any one of claims 1 to 9 is implemented.

Images