Image stitching alignment point verification method, device, equipment and storage medium

Abstract

The present application discloses an image stitching alignment point verification method, device, equipment and storage medium, which relate to the field of image recognition. Two adjacent scanned images containing overlapping areas are obtained, target pixel points are respectively selected in the overlapping areas, and corresponding point-selected images are generated according to the target pixel points; all original contours contained in the frame line image are identified and mapped to determine the same target contour; based on the center of the selected target contour as the origin, the target contour is framed to generate a target framed image, and the adjacent scanned images are spliced ​​and fused to generate a panoramic image. This scheme adopts the method of first selecting to determine the point-selected image, and then generating the target framed image based on the center point of the target contour in the point-selected image, which can ensure the consistency of the picture content and the accuracy of splicing and fusion.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image recognition, and in particular to a method, device, equipment and storage medium for checking image stitching alignment points. Background Art

[0002] When using visual defect detection technology to detect appearance defects of target objects, the camera cannot directly obtain a panoramic image of the target object to be detected due to the size limitation of the target object. Therefore, the industry usually uses image stitching, that is, the camera scans the target object one by one, stitching the images generated by multiple areas of the target object to obtain a complete panoramic image.

[0003] In the related technologies, the image stitching methods all involve the stitching points (features) of the images, that is, the corresponding stitching points are selected from the two images to be stitched, and the stitching points are overlapped to achieve the image stitching. However, this technology uses a box selection method when selecting the stitching points, and inevitably there will be a situation where the stitching points are not neatly selected. The puzzle points need to be verified. If the verification fails, the stitching points need to be re-selected to determine the stitching points, which ultimately reduces the efficiency of puzzle fusion. Summary of the invention

[0004] The embodiments of the present application provide an image stitching alignment point verification method, device, equipment and storage medium to solve the problem that the stitching point boxes of the puzzle are not aligned, resulting in reduced puzzle fusion efficiency.

[0005] On the one hand, the present application provides an image stitching alignment point verification method, the method comprising:

[0006] Acquire two adjacent scanned images containing overlapping areas, select target pixels in the overlapping areas respectively, and generate corresponding point-selected images according to the target pixels; the two point-selected images contain at least some of the same pixel content;

[0007] Identifying all component contours contained in the point-selected image and mapping the component contours to determine the same target contour;

[0008] Based on the center of the selected target contour being the alignment point, the target contour is framed to generate a target framed image, and the adjacent scanned images are spliced ​​and fused to generate a panoramic image.

[0009] Specifically, generating a corresponding point-selected image according to the target pixel point includes:

[0010] The target pixel points selected in the scanned image are respectively used as the center coordinates of the frame selection, and the selected image is framed in the scanned image with a set target frame selection size;

[0011] The target pixel points selected in the two scanned images are located in the same size range, and at least one of the component outlines in the selected images respectively selected are the same.

[0012] Specifically, identifying all component contours contained in the point-selected image and mapping the component contours includes:

[0013] Recognize the selected images respectively, and extract the component outlines and corresponding coordinate information contained therein;

[0014] The complete component contours in the two point-selected images are extracted respectively, and the coordinate information and contour lines of each component contour are mapped to determine the matching component contours.

[0015] Specifically, determining the same target profile includes:

[0016] Generate contour matching priorities for matching component contours according to size and shape; the contour matching priority is positively correlated with the contour size and shape complexity;

[0017] The component contours with the highest priority and the same in the two point selection images are determined as the target contours.

[0018] Specifically, based on the selected center of the target outline as the origin, the target outline is framed to generate a target framed image, including:

[0019] In the two scanned images, the geometric center point is determined according to the shape and size of the target contour respectively, and the geometric center is used as the alignment point to generate the target frame selection image by expanding the target size.

[0020] Specifically, at least two target pixel points are selected in the overlapping area of ​​each scanned image, and corresponding target framed images are formed respectively, and all target framed images in each scanned image form a spliced ​​image group.

[0021] Specifically, determining the same target profile includes:

[0022] Selecting a target number of complete and adjacent component contours in the point selection image to form the target contour, and determining the size of the target contour;

[0023] The geometric center point is determined according to the shape and size of the target outline, and the target frame image is generated by expanding the target size; the target frame image is used as an image alignment block, and the alignment block is used for alignment and fusion during stitching and fusion to generate a panoramic image.

[0024] On the other hand, the present application provides an image stitching alignment point verification device, the device comprising:

[0025] The first frame selection module is used to obtain two adjacent scanned images containing overlapping areas, select target pixels in the overlapping areas respectively, and generate corresponding point-selected images according to the target pixels; the two point-selected images contain at least part of the same pixel content;

[0026] A determination module, used for identifying all component contours contained in the point selection image and mapping the component contours to determine the same target contour;

[0027] The second frame selection module is used to frame the target outline based on the center of the selected target outline as the alignment point to generate a target point selection image, and to splice and fuse the adjacent scanned images to generate a panoramic image.

[0028] 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 is loaded and executed by the processor to implement the image stitching alignment point verification method described in the above aspects.

[0029] On the other hand, the present application provides a computer-readable storage medium, which stores at least one instruction, at least one program, a code set or an instruction set. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the image stitching alignment point verification method described in the above aspects.

[0030] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0031] This application abandons the traditional mechanism of automatic identification of matching alignment points and manual selection, and instead adopts a fuzzy preliminary selection mode of selecting target pixels. This selection mode does not consider the accuracy of the picture content. The picture contour in the preliminary selection image is subsequently identified and mapped, and the target contour is determined from it, and then the fine recognition step is entered. The method of regenerating the target selection image with the center point and contour size of the target contour can ensure the consistency of the picture content and the accuracy of splicing and fusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a schematic diagram of stitching a panoramic image in the related art;

[0033] Figure 2 is a flow chart of the image stitching alignment point verification method provided by an embodiment of the present application;

[0034] Figure 3 This is a schematic diagram of selecting a target pixel point on a scanned image provided by an embodiment of the present application;

[0035] Figure 4 yes Figure 3 An enlarged schematic diagram of the point-selected image generated in FIG.

[0036] Figure 5 It is a schematic diagram of generating a target frame selection diagram according to alignment points provided in an embodiment of the present application;

[0037] Figure 6 A structural block diagram of a computer device provided by an exemplary embodiment of the present application is shown;

[0038] Figure 7 A schematic diagram showing a target profile combined with several adjacent element profiles;

[0039] Figure 8 A structural block diagram of an image stitching alignment point verification device provided by an embodiment of the present application is shown;

[0040] Fig. 9 A structural block diagram of a computer device provided by an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0041] 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.

[0042] 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.

[0043] Figure 1It is a schematic diagram of stitching panoramic images in the related art. After selecting a certain target contour (or Mark point), an area needs to be framed, that is, the a area and the a' area in Figure A and Figure B. In theory, the a area and the a' area should correspond to the same area range of the target to be measured (material board), but the present application is for the material panoramic image in the semiconductor field, and the panoramic image is generated by scanning the material board one-to-one. As a master drawing, it is necessary to ensure pixel-level accuracy, and the Mark point and the component contour are much larger than the unit pixel particle size. Under the requirements of large size and high precision, the lying image scanned by the line scan camera contains a large number of identical and similar components, and the AI ​​frame selection operation is not applicable. Especially in the case where the object is offset during the scanning process of two adjacent trips, the overall B image is relatively tilted with the A image, and even if the coordinate range mapping frame selection method is adopted, it cannot be guaranteed to be completely consistent. Even if the manual operation method is adopted, it cannot be guaranteed that the size of the a area and the a' area is completely consistent with the framed area. The left a area only contains one target contour, and the adjacent contour curve is also framed in the right a' area. Similarly, the target outline of area b in picture A is relatively located in the center of the framed area, but the target outline of area b' in picture B is biased upward. Such spliced ​​images cannot be directly fused, and multiple subsequent verifications and frame selections and cropping are required, which will lead to low fusion efficiency.

[0044] Figure 2 : is a flow chart of the image stitching alignment point verification method provided by the embodiment of the present application, comprising the following steps:

[0045] S1, obtaining two adjacent scanned images containing overlapping areas, selecting target pixels in the overlapping areas respectively, and generating corresponding selected images according to the target pixels;

[0046] Large-size panoramic images are usually stitched together after multiple scans by a line scan camera. Therefore, an overlapping area of ​​a certain width needs to be set between two adjacent scanned images. The stitching points or stitching blocks are extracted from this overlapping area.

[0047] Figure 3 It is a schematic diagram of selecting target pixel points in a scan image provided by an embodiment of the present application. The present application abandons the traditional AI recognition and manual frame selection of contour images, but manually selects pixel points first. Area a in the A scan image and area b in the B scan image correspond to the same overlapping area. This solution is to automatically frame selected images from area a and area b according to pixel points, and then continue to generate framed images based on the point selected images.

[0048] In this step, the point selection operation is based on manual operation or machine random operation. As long as the selection is in the overlapping area, the point selection is in pixels. After selecting a target pixel in the A-scan image, continue to select the corresponding pixel from the B-scan image.

[0049] Figure 4 yes Figure 3 The enlarged schematic diagram of generating the point-selected image in the figure, the target pixel point B is not required to be completely consistent with the A-scan pixel point, as long as it is within a reasonable range. This process can usually select two target pixel points through visual mapping or computer automatic mapping. After selecting two target pixel points, the point-selected image can be automatically generated according to the target pixel points. The size of the two point-selected images can be the same or different. In theory, as long as there is an intersection between the two, it can be used for subsequent contour mapping from the intersection.

[0050] S2, identifying all component contours contained in the point-selected image and mapping the component contours to determine the same target contour;

[0051] Since there are many components such as metal surfaces and circuits on the material board, even the point selection image generated after clicking will contain many component contours. In this case, matching and mapping are required, that is, finding the component contours of the intersection. In the embodiment of the present application, it is necessary to perform contour recognition, contour mapping and matching on two point selection images, and select the target contour of at least one target component from the matching consistent contours. The target contour is used to subsequently determine the alignment point and generate the target frame selection image.

[0052] S3, based on the center of the selected target contour as the alignment point, the target contour is framed to generate a target framed image, and adjacent scanned images are spliced ​​and fused to generate a panoramic image.

[0053] After determining at least one target contour, a geometric analysis method can be used to identify the geometric center point of the contour, and the geometric center point can be determined as the alignment point. Further, the target framed image, that is, the image block used for alignment verification, is automatically framed based on the alignment point and the target contour.

[0054] Figure 5 is a schematic diagram of generating a target frame selection image according to an alignment point provided in an embodiment of the present application. Assuming that a triangular outline is selected as the target outline after mapping and matching in the first point selection image 510 and the second point selection image 520, the corresponding geometric center is identified by the size of the triangular outline in the two images, that is, Figure 5 The red dot marked in the figure is used as the alignment point for identification and confirmation. It can be seen that the alignment point identified by this method has higher accuracy than the traditional method of selecting mark identification or target component positioning, and the point is not limited to the iconic contour line.

[0055] Furthermore, after the alignment point is determined, a target frame selection image is generated based on the alignment point and the target contour, that is, Figure 5 The first target frame selection image 511 and the second target frame selection image 521 in the figure. It should be noted here that the first target frame selection image 511 and the second target frame selection image 521 are exactly the same, and the corresponding areas on the material board are completely overlapped, which is completely different from the point selection image, and there is no need to worry about the problem of inconsistent overlap.

[0056] In particular, in some special scenarios, when the pixel content of the B scan image is tilted due to shaking or other vibrations of the material board, the selected image is framed according to the angle of the scan image frame, and its picture pixels are also tilted. However, the method of using contour mapping to determine the alignment point can accurately identify the center of the contour. The angle and picture of the target framed image 511 generated based on the target contour are normal, and are completely consistent with the pixel content of the target framed image 521 in the A image, which also ensures the accuracy of splicing and fusion.

[0057] Based on the above rules, at least two target pixels need to be selected in the overlapping area of ​​the two scanned images, and the corresponding target framed images are determined. Finally, a stitched image group is formed based on all the target framed images in each scanned image, and alignment and stitching fusion are performed according to the stitched image group to finally generate a complete panoramic image.

[0058] In summary, this application abandons the traditional mechanism of automatic identification of matching alignment points and manual selection, and instead adopts a fuzzy preliminary selection mode of selecting target pixels. This selection mode does not consider the accuracy of the picture content. The picture contour in the preliminary selection image is subsequently identified and mapped, and the target contour is determined, and then the fine recognition step is entered. The method of regenerating the target selection image with the center point and contour size of the target contour can ensure the consistency of the picture content and the accuracy of splicing and fusion.

[0059] Based on the above content, it can be known that when selecting the target pixel points, machine recognition can be used. As long as the positions of the two target pixel points are maintained within a certain accuracy range, for manual annotation, the size of the automatically generated point selection image can be increased. In this way, at least one of the same component contours in the selected point selection images can also ensure that the mapping process has sufficient selection. When generating the point selection image, the target pixel points are used as the center coordinates of the selection, and the point selection image is selected in the scanned image with the set target selection size.

[0060] In some embodiments, the steps of identifying all component outlines contained in the point-selected image and mapping the component outlines may include the following steps:

[0061] A. Recognize the selected images respectively and extract the component outlines and corresponding coordinate information contained therein;

[0062] B. Extract the complete component contours in the two point-selected images respectively, map them according to the coordinate information and contour lines of each component contour, and determine the matching component contours.

[0063] The coordinate information extracted in this step is based on the scanned image. Figure 4 and Figure 5 As can be seen in the figure, the material board contains a large number of gold surface contours of the same shape and size. Simple contour recognition and matching may result in misalignment, so coordinate information is required for verification. The incomplete contour lines intercepted in the point selection diagram are directly discarded, and only the complete component contour is mapped and matched.

[0064] In some embodiments, the process of determining the same target profile may be further implemented as follows:

[0065] a. Generate contour matching priorities for matching component contours according to size and shape; the contour matching priority is positively correlated with the contour size and shape complexity;

[0066] b. Determine the component outline with the highest priority and the same in the two point selection images as the target outline.

[0067] For example Figure 5 In the left and right mapping relationships of the two point selection images, both rectangular and circular contours can be mapped consistently. When selecting the target contour, irregular or non-circular contours are given priority. Figure 5 As can be seen from the figure, circular contours cannot guarantee the consistency of the selection when offset, so shapes with larger sizes and curved contours are easier to locate and generate target selection images. Therefore, contours with larger sizes, more obvious curved shapes, and more complex contours have higher priorities. Figure 5 The outline of the triangle is selected as the target outline. After the target outline is selected, the geometric center point is determined in the two scanned images according to the shape and size of the target outline, and the geometric center is used as the alignment point to generate a target frame selection image by expanding the target size (for example, the size of the expanded complete target outline). Of course, in some embodiments, for the convenience of observation, after determining the alignment point, the target frame selection image can be expanded according to the set size. In this case, the target outline image may not include the complete target outline, or may additionally include the target outline and the outline of other components.

[0068] In some other embodiments, in order to ensure the accuracy of the frame selection image, a superimposed outline may be selected. Figure 6A schematic diagram of determining the target contour is shown in the figure, in which the shape of the triangular element contour is superimposed on the rectangular gold surface, and the center point of the two structures is used as the alignment point for higher accuracy.

[0069] In some other embodiments, an alignment block may be selected to upgrade the alignment points, that is, a target number of complete and adjacent component contours in the point selection image are selected to form a target contour, and the size of the target contour is determined. Figure 7 A schematic diagram showing a target profile combined with several adjacent element profiles.

[0070] Furthermore, the geometric center point is determined according to the shape and size of the target outline, and the target frame image is generated by expanding the target size. The target frame image here is used as an image alignment block, and the alignment block is used for alignment and fusion to generate a panoramic image during stitching and fusion. Different from the previous process, the concept of alignment point is not considered in this step, but the entire frame image is used as the alignment image, which has higher alignment accuracy.

[0071] Figure 8 The structural block diagram of the image stitching alignment point verification device provided by the embodiment of the present application is shown, and the device includes:

[0072] The first frame selection module 810 is used to obtain two adjacent scanned images containing overlapping areas, select target pixels in the overlapping areas respectively, and generate corresponding point-selected images according to the target pixels; the two point-selected images contain at least part of the same pixel content;

[0073] A determination module 820, configured to identify all component contours contained in the point-selected image and map the component contours to determine the same target contour;

[0074] The second frame selection module 830 is used to frame the target outline to generate a target point selection image based on the center of the selected target outline as the alignment point, and to splice and fuse the adjacent scanned images to generate a panoramic image.

[0075] The image stitching alignment point verification device provided in the embodiment of the present application can be applied to the image stitching alignment point verification method provided in the above embodiment. For relevant details, refer to the above method embodiment. The implementation principle and technical effect are similar and will not be repeated here.

[0076] It should be noted that the image stitching alignment point verification device provided in the embodiment 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 image stitching alignment point verification 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 image stitching alignment point verification method provided in the above method embodiment and the implementation method of the image stitching alignment point verification device provided in this embodiment belong to the same concept. The specific implementation process of the image stitching alignment point verification device provided in this embodiment is detailed in the above method embodiment, which will not be repeated here.

[0077] Fig. 9 The block diagram of a computer device provided by an exemplary embodiment of the present application is shown. The computer device may be 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. The processor and the memory may be connected by a bus or otherwise. 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.

[0078] 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.

[0079] 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.

[0080] 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.

[0081] 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.

[0082] 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.

[0083] 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.

[0084] 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.

[0085] 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.

[0086] 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 method for checking image stitching alignment points, characterized in that: The method comprises: Acquire two adjacent scanned images containing overlapping areas, select target pixels in the overlapping areas respectively, and generate corresponding point-selected images according to the target pixels; the two point-selected images contain at least some of the same pixel content; Identifying all component contours contained in the point-selected image and mapping the component contours to determine the same target contour; Based on the center of the selected target contour being the alignment point, the target contour is framed to generate a target framed image, and the adjacent scanned images are spliced ​​and fused to generate a panoramic image.

2. The method according to claim 1, characterized in that: The step of generating a corresponding point-selected image according to the target pixel point comprises: The target pixel points selected in the scanned image are respectively used as the center coordinates of the frame selection, and the selected image is framed in the scanned image with a set target frame selection size; The target pixel points selected in the two scanned images are located in the same size range, and at least one of the component outlines in the selected images respectively selected are the same.

3. The method according to claim 2, characterized in that Identifying all component outlines contained in the point-selected image and mapping the component outlines includes: Recognize the selected images respectively, and extract the component outlines and corresponding coordinate information contained therein; The complete component contours in the two point-selected images are extracted respectively, and the coordinate information and contour lines of each component contour are mapped to determine the matching component contours.

4. The method according to claim 3, characterized in that The determining of the same target profile comprises: Generate contour matching priorities for matching component contours according to size and shape; the contour matching priority is positively correlated with the contour size and shape complexity; The component contours with the highest priority and the same in the two point selection images are determined as the target contours.

5. The method according to any one of claims 1 to 4, characterized in that: Based on the selected center of the target outline as the origin, the target outline is framed to generate a target framed image, including: In the two scanned images, the geometric center point is determined according to the shape and size of the target contour respectively, and the geometric center is used as the alignment point to generate the target frame selection image by expanding the target size.

6. The method according to claim 5, characterized in that At least two target pixel points are selected in the overlapping area of ​​each scanned image, and corresponding target framed images are formed respectively. All target framed images in each scanned image form a spliced ​​image group.

7. The method according to claim 4, characterized in that The determining of the same target profile comprises: Selecting a target number of complete and adjacent component contours in the point selection image to form the target contour, and determining the size of the target contour; The geometric center point is determined according to the shape and size of the target outline, and the target frame image is generated by expanding the target size; the target frame image is used as an image alignment block, and the alignment block is used for alignment and fusion during stitching and fusion to generate a panoramic image.

8. An image stitching alignment point verification device, characterized in that: The device comprises: The first frame selection module is used to obtain two adjacent scanned images containing overlapping areas, select target pixels in the overlapping areas respectively, and generate corresponding point-selected images according to the target pixels; the two point-selected images contain at least part of the same pixel content; A determination module, used for identifying all component contours contained in the point selection image and mapping the component contours to determine the same target contour; The second frame selection module is used to frame the target outline based on the center of the selected target outline as the alignment point to generate a target point selection image, and to splice and fuse the adjacent scanned images to generate a panoramic image.

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 image stitching alignment point verification 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 is loaded and executed by the processor to implement the image stitching alignment point verification method as described in any one of claims 1 to 7.

Images