Map matching method and device and storage medium

Abstract

Embodiments of the invention provide a map matching method and apparatus, and a storage medium. The method comprises the steps of obtaining a to-be-matched map; inputting a to-be-matched map into the image segmentation model to obtain a plurality of image segmentation mask images; determining a plurality of target mask images greater than a threshold value from the plurality of image segmentation mask images; performing positive and negative binarization processing on each target mask image to obtain a positive and negative binarized mask image; aiming at each standard map in a preset standard map library, obtaining a standard map mask map and the size of the standard map mask map; generating a new positive and negative binarization mask picture of each target mask picture according to the standard map mask picture and the size; based on an image matching model, respectively matching the generated new positive and negative binarization mask images with a standard map mask image, and respectively outputting collocation point pairs of each target mask image and the standard map mask image; and determining a standard map matched with the to-be-matched map according to the maximum matching point pairs. According to the invention, intelligent map matching can be realized.

Description

Technical Field

[0001] This document relates to the field of image processing technology, and in particular to a map matching method, device and storage medium. Background Art

[0002] As an abstract, symbolic and visual description of geographic spatial information, the importance of maps cannot be ignored. With the development of informatization, the use of maps has penetrated into all aspects of daily life, from navigation, takeaway ordering to address query, etc., maps provide a variety of convenient services.

[0003] The prior art mainly utilizes deep learning technology to identify the map to be identified, determine the features of the map to be identified, find similar maps based on the features of the map to be identified, and thus determine the geographic location information of the map to be identified.

[0004] However, compared with the recognition of ordinary pictures, there are many geographical details and texture features on the map, and the texture features, production standards and complexity of different maps are different and overlap with each other. When using deep learning to recognize maps, there is a lot of noise and the feature extraction accuracy is low, making it difficult to quickly and accurately recommend standard map references for the map to be matched. Summary of the invention

[0005] In view of the above solutions, the present application aims to propose a map matching method, device and storage medium to solve at least one of the above technical problems.

[0006] In a first aspect, one or more embodiments of this specification provide a map matching method, including:

[0007] S1. Obtain the map to be matched;

[0008] S2, inputting the map to be matched into an image segmentation model to obtain a plurality of image segmentation mask maps;

[0009] S3, determining a plurality of target mask images greater than a threshold value from the plurality of image segmentation mask images;

[0010] S4, performing positive and negative binarization processing on each of the target mask images to obtain a positive binarization mask image and a negative binarization mask image;

[0011] S5. For each standard map in the preset standard map library, obtain a standard map mask image and a size of the standard map mask image;

[0012] S6. For each standard map in the preset standard map library, generate a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size; wherein the size represents the number of pixels in height and the number of pixels in width;

[0013] S7. Based on the image matching model, the generated new positive binary mask image and the new debinarized mask image are matched with the standard map mask image respectively, and the first matching point pair and the second matching point pair of each target mask image and the standard map mask image are output respectively, wherein the first matching point pair represents the matching point pair of the new positive binary mask image of each target mask image in the map to be matched and the standard map mask image, and the second matching point pair represents the matching point pair of the new debinarized mask image of each target mask image in the map to be matched and the standard map mask image; and

[0014] S8. Determine a standard map that matches the map to be matched according to the first matching point pairs and the second matching point pairs of each of the target mask images.

[0015] Furthermore, the step of inputting the to-be-matched map into an image segmentation model to obtain a plurality of image segmentation mask maps comprises:

[0016] Dividing the to-be-matched map into grids to generate a plurality of prompt points;

[0017] Based on the image segmentation model, using multiple prompt points, extracting the image texture and tile features of the map to be matched; and

[0018] According to the image texture and the image block features, the map to be matched is segmented into a plurality of image segmentation mask images.

[0019] Further, the determining a plurality of target mask images greater than a threshold value according to the plurality of image segmentation mask images comprises:

[0020] Determining the area of ​​each of the image segmentation mask images;

[0021] sorting the image segmentation mask images according to the areas; and

[0022] The image segmentation mask image that is greater than a threshold is selected as the target mask image.

[0023] Furthermore, for each standard map in the preset standard map library, generating a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size includes:

[0024] If the width and height of the standard map mask are different,

[0025] According to the standard map mask, the number of pixels of the height of the standard map mask and the number of pixels of the width of the standard map mask are obtained;

[0026] Determine the minimum number of pixels between the number of pixels of the height and the number of pixels of the width in the standard map mask;

[0027] The number of pixels of each target mask image is obtained respectively; wherein the number of pixels of the target mask image includes the number of pixels of the height of the target mask image and the number of pixels of the width of the target mask image;

[0028] Determining a first scale adjustment benchmark according to the minimum number of pixels in the standard map mask image, wherein the first scale adjustment benchmark is the height or width of the standard map mask image;

[0029] Determining a first adjustment ratio according to the first ratio adjustment reference and the number of pixels of each of the target mask images;

[0030] According to the first adjustment ratio, the number of pixels in width and the number of pixels in height of the positive binary mask image and the debinarized mask image of each of the target mask images are respectively enlarged or reduced in the same ratio to generate a new positive binary mask image and a new debinarized mask image;

[0031] If the width and height of the standard map mask are the same,

[0032] According to the target mask image, the number of pixels of the height of the target mask image and the number of pixels of the width of the target mask image are obtained;

[0033] Determine the minimum pixel between the number of height pixels and the number of width pixels in the target mask image;

[0034] Determining a second ratio adjustment reference according to the minimum number of pixels in the target mask image, where the second ratio adjustment reference is the height or width in the target mask image;

[0035] Determining a second adjustment ratio according to the second ratio adjustment reference and the number of pixels of each standard map mask image;

[0036] According to the second adjustment ratio, the width pixel number and the height pixel number of the positive binary mask image and the debinarized mask image of each of the target mask images are respectively enlarged or reduced in the same proportion to generate a new positive binary mask image and a new debinarized mask image.

[0037] Further, the method of matching the generated new positive binary mask image and the generated new inverse binary mask image with the standard map mask image respectively based on the image matching model to output the first matching point pair and the second matching point pair of each target mask image and the standard map mask image respectively includes: the image matching model includes a feature detector and a graph neural network model, and

[0038] Using a feature detector, extracting point and line segment features of the new positive binarization mask image, the new debinarization mask image, and the standard map mask image; and

[0039] Based on the graph neural network model, the point and line segment features of the new positive binary mask image and the new debinarized mask image are matched with the point and line segment features of the standard map mask image, and the first matching point pair and the second matching point pair of the new positive binary mask image and the new debinarized mask image and the standard map mask image are output respectively.

[0040] Furthermore, the preset standard map library includes a plurality of standard maps; each of the standard maps is binarized to obtain a plurality of standard map mask images; and

[0041] Based on the image matching model, the generated new positive binary mask image and the new inverse binary mask image are matched with the standard map mask image respectively, and the first matching point pair and the second matching point pair of each target mask image and the standard map mask image are output respectively, including:

[0042] For each standard map mask, determining the size of the current standard map mask;

[0043] According to the size of the current standard map mask image, the sizes of the positive binary mask image and the debinarized mask image are updated to obtain the new positive binary mask image and the new debinarized mask image;

[0044] Determining the first matching point pair according to the current standard map mask image and the new positive binary mask image; and

[0045] The second matching point pair is determined according to the current standard map mask image and the new debinarized mask image.

[0046] Further, the determining a standard map matching the map to be matched according to the first matching point pairs and the second matching point pairs of each of the target mask images comprises:

[0047] The first matching point pairs and the second matching point pairs of each of the standard map mask images and each of the target mask images are sorted respectively, and a standard map having the most matching point pairs when matching with the target mask image is output.

[0048] Furthermore, a distributed computing framework is used to execute step S5 and steps S6 to S8.

[0049] In a second aspect, an embodiment of the present application provides a map matching device, including:

[0050] A first acquisition module is used to acquire a map to be matched;

[0051] A first determination module is used to input the map to be matched into an image segmentation model to obtain a plurality of image segmentation mask maps;

[0052] A second determination module is used to determine a plurality of target mask images greater than a threshold from the plurality of image segmentation mask images;

[0053] A third determination module is used to perform positive and negative binarization processing on each of the target mask images to obtain a positive binarization mask image and a negative binarization mask image;

[0054] A second acquisition module is used to acquire a standard map mask image and a size of the standard map mask image for each standard map in a preset standard map library;

[0055] A fourth determination module is used to generate a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size; wherein the size represents the number of pixels in height and the number of pixels in width; and

[0056] a fifth determination module, for matching the generated new positive binary mask image and the new debinarized mask image with the standard map mask image respectively based on the image matching model, and outputting first matching point pairs and second matching point pairs of each target mask image and the standard map mask image respectively, wherein the first matching point pairs represent matching point pairs of the new positive binary mask image of each target mask image in the map to be matched and the standard map mask image, and the second matching point pairs represent matching point pairs of the new debinarized mask image of each target mask image in the map to be matched and the standard map mask image; and

[0057] The sixth determination module is used to determine a standard map that matches the map to be matched according to the first matching point pairs and the second matching point pairs of each of the target mask images.

[0058] In a third aspect, an embodiment of the present application provides a storage medium for storing computer-executable instructions, characterized in that when the computer-executable instructions are executed, the steps of the map matching method described in any one of the first aspects may be performed.

[0059] Compared with the prior art, this application can at least achieve the following technical effects:

[0060] This application uses an image segmentation model and a map to be matched to divide a map into multiple parts, and then quickly find the target area of ​​interest to the user from these maps, thereby reducing the noise when using the map and improving the search speed; then use positive and negative binarization technology to process the image from different angles, highlighting different edges and textures in the image, so as to facilitate subsequent map matching; then modify the size of the positive and negative binary mask image to make it consistent with the standard map mask. Figure 1 The method is to reduce the matching error of the matching point pairs in the mapping process; then use the image matching model to output the matching point pairs of the positive and negative binarization mask map and the standard map mask map respectively, and ensure as many matching point pairs as possible through the above-mentioned positive and negative binarization and size modification; then use the sorting to quickly find the standard map that matches the map to be matched, and improve the matching accuracy and efficiency of the map through the above-mentioned positive and negative binarization technology, map size modification and other means. In summary, by performing positive and negative binarization processing on the map to be processed, modifying the size of the map, and comparing the modified size map with the standard map, the most matching standard map can be accurately found from each standard map. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0062] Figure 1 A flow chart of a map matching method provided for one or more embodiments of this specification;

[0063] Figure 2 A schematic diagram of a map segmentation process of a map matching method provided in one or more embodiments of this specification;

[0064] Figure 3 A schematic diagram of the forward and reverse binarization processing process of a map matching method provided in one or more embodiments of this specification;

[0065] Figure 4 A schematic diagram of matching a positive binary mask image with a standard map mask image provided in one or more embodiments of this specification;

[0066] Figure 5 A schematic diagram of matching a debinarized mask image with a standard map mask image provided in one or more embodiments of this specification;

[0067] Figure 6A schematic diagram of the structure of a map matching device provided in one or more embodiments of this specification. DETAILED DESCRIPTION

[0068] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0069] Image matching aims to identify and align the content or structure with the same / similar attributes in two images at the pixel level. In the current context of Internet big data, image matching of maps can be used to identify the area represented by the map, compare the differences in the main details of the map images, and thus dig out potential problematic map images on the Internet. It is of great significance and role in the development of the geographic information industry and the supervision of Internet maps. With the rapid development of the Internet, various maps are circulated on the Internet. If the maps are not unified, it will affect people's cognitive errors, thus affecting normal social and economic activities.

[0070] In order to solve the technical problems existing in the prior art, this application proposes a map matching method, such as Figure 1 As shown, the specific steps are as follows:

[0071] Step S1, obtaining a map to be matched.

[0072] In an embodiment of the present application, the system determines and obtains a map to be matched based on a map uploaded by a user.

[0073] Step S2: input the map to be matched into an image segmentation model to obtain a plurality of image segmentation mask maps.

[0074] In the embodiment of the present application, based on the image segmentation model (SAM, Segment Anything Model), texture features and block features on the map image to be matched are extracted, and the map image to be matched is segmented to generate multiple image segmentation mask images.

[0075] Specifically, the SAM image segmentation model is trained and constructed using a large-scale image dataset of 11 million images containing 1 billion masks. On unknown datasets and tasks, SAM has good zero-trigger segmentation performance and can solve a wide range of segmentation tasks. Its segmentation process is driven by different cues such as points or boxes to indicate the pixel semantics and regional location of the target object. This application divides the map image into uniform grids, uses the grid center points as cue points, generates multiple cue points for the SAM model, and then inputs them into the SAM model to segment the map image and generate image segmentation mask maps for multiple targets. The uniform grid division cue point generation strategy is then used to improve the segmentation coverage of map targets, such as Figure 2 shown.

[0076] For example: there are areas such as Beijing, Hebei, and Chongli on the map image to be identified. Each area is distinguished by lines. The SAM image segmentation model is used to segment the map according to the texture and block features of the map to be matched, and the boundary maps of the image segmentation mask maps of Beijing, Hebei, Chongli and other areas are obtained respectively.

[0077] Step S3: determining a plurality of target mask images that are greater than a threshold value from the plurality of image segmentation mask images.

[0078] In an embodiment of the present application, each image segmentation mask map is traversed, the size of each image segmentation mask map is obtained, and the area of ​​each image segmentation mask map is calculated. The image segmentation mask maps are sorted according to the area, and a target mask map greater than a threshold is selected.

[0079] For example, there are image segmentation masks: Picture 1, Picture 2, Picture 3, Picture 4. Use an image processing library (such as PIL or OpenCV) to open each image segmentation mask and read its size (width and height). Picture 1 has a width of 300 pixels and a height of 200 pixels; Picture 2 has a width of 800 pixels and a height of 600 pixels; Picture 3 has a width of 1920 pixels and a height of 1080 pixels; Picture 4 has a width of 5000 pixels and a height of 3333 pixels. Then calculate the area based on the width and height of each picture. The area of ​​Picture 1 is 60,000 pixels; the area of ​​Picture 2 is 480,000 pixels, the area of ​​Picture 3 is 2,073,600 pixels, and the area of ​​Picture 4 is 16,665,000 pixels. Sort the pictures by area: Picture 4>Picture 3>Picture 2>Picture 1. If the threshold is 48,000 pixels, then Pictures 4, 3, and 2 are determined to be possible target masks.

[0080] Step S4, performing positive and negative binarization processing on each of the target mask images to obtain a positive binarization mask image and a negative binarization mask image.

[0081] In an embodiment of the present application, the target object is binarized to 0 and 1, the inside of the target object is assigned a value of 1 or the outside of the target object is assigned a value of 1, and a binarization conversion is performed to generate two different binary images, respectively obtaining a positive binarization mask image and a negative binarization mask image.

[0082] Specifically, the target segmentation mask image is first binarized, and then the binarized mask image is debinarized. In addition, the resolutions of different map images are different, resulting in different details of the SAM model segmentation boundary information in the same area, such as Figure 3 As shown in Figure 2, if the map object segmentation boundary is smooth and curved to different degrees, it will greatly limit the matching effect of the map boundary.

[0083] For example: first find the contour of the target segmentation mask image, and obtain a positive binary mask image by assigning the inside of the contour to 1 (white) and the outside of the contour to 0 (black); then obtain a de-binarized mask image by assigning the inside of the contour to 0 (black) and the outside of the contour to 1 (white). The positive binary mask image and the de-binarized mask image have different degrees of smoothness and curvature relative to the boundary concern, and different matching results will be recommended, thereby improving the accuracy of map matching.

[0084] Step S5: for each standard map in the preset standard map library, obtain a standard map mask image and the size of the standard map mask image.

[0085] In an embodiment of the present application, each standard map in the standard map library is traversed, wherein the maps in the preset standard map library are also based on the image segmentation tool SAM, each object of the standard map is segmented, and the corresponding mask map is obtained, and then the mask map is binarized to obtain the standard map mask map; and then the pixel size of the standard map mask map is determined.

[0086] Specifically, the preset standard map library includes standard maps and standard map mask maps, wherein the standard map mask map is obtained by segmenting each standard map using the SAM image segmentation model to obtain multiple image segmentation mask maps; and then binarizing the multiple image segmentation mask maps to obtain a standard map mask map.

[0087] Step S6, generating a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size; wherein the size represents the number of pixels in height and the number of pixels in width.

[0088] In the embodiment of the present application, the pixel size of each standard map mask image, the pixel size of the positive binary mask image, and the pixel size of the debinarized mask image are read, and the pixel sizes of the positive binary mask image and the debinarized mask image are adjusted to the same size as the standard map mask image while maintaining their aspect ratio. Figure 1That is, according to the standard map mask map (binarized mask map) of the standard map library, the positive binary mask map and the inverse binary mask map to be matched are aligned according to the standard map library. Figure 2 The minimum length and width of the binary mask image is proportionally reduced or enlarged to generate a multi-scale binary mask image, thereby ensuring that the new positive binary mask image and the new inverse binary mask image to be matched are consistent with the standard map. Figure 2 The scale consistency of the binary mask image is calculated to generate the final multi-target binary mask image to be matched.

[0089] For example: The preset standard map library includes standard map masks Figure 1 , Standard map mask Figure 2 , Standard map mask Figure 3 etc. Standard map masks Figure 1 The width is 300 pixels and the height is 200 pixels; standard map mask Figure 2 The width is 400 pixels and the height is 300 pixels; standard map mask Figure 3 The width is 100 pixels and the height is 100 pixels.

[0090] The width of the positive and negative binary mask is 100 pixels and the height is 200 pixels; determine the standard map mask Figure 1 , Standard map mask Figure 2 , Standard map mask Figure 3 The minimum pixel height or width in the pixel size of the standard map mask Figure 1 , Standard map mask Figure 2 , Standard map mask Figure 3 The pixel sizes of the positive binary mask image and the negative binary mask image are adjusted according to the ratio corresponding to the minimum pixel size of the image. The adjusted new positive and negative binary mask images have the same proportional sizes as the standard map mask image, with sizes of 100 pixels in width and 200 pixels in height; 150 pixels in width and 300 pixels in height; and 100 pixels in width and 200 pixels in height, respectively, for subsequent recognition.

[0091] The calculation process is: when the width and height pixels of the standard map mask are the same, the standard map mask Figure 3 The width is 100 pixels and the height is 100 pixels; when the width of the positive and negative binary mask images is 100 pixels and the height is 200 pixels, the minimum number of pixels in the positive and negative binary mask images is determined to be the width. According to the number of pixels in the width of the positive and negative binary mask images and the number of pixels in the width of the standard map mask image, the adjustment ratio is determined to be 1. Therefore, the adjusted new positive and negative binary mask images have the same proportional size as the standard map mask image, with a width of 150 pixels and a height of 300 pixels respectively.

[0092] When the width and height of the standard map mask are different, Figure 2 The width of the standard map mask is 300 pixels and the height is 200 pixels; the width of the positive and negative binary mask is 100 pixels and the height is 200 pixels. Figure 1 The minimum number of pixels in is the height, according to the standard map mask Figure 1 The number of height pixels and the number of height pixels of the positive and negative binary mask images are determined to be adjusted to 1 times, so the adjusted new positive and negative binary mask images have the same proportional size as the standard map mask image, and the size is still 100 pixels wide and 200 pixels high.

[0093] Step S7, based on the image matching model, match the generated new positive binary mask image and the new debinarized mask image with the standard map mask image respectively, and output the first matching point pair and the second matching point pair of each target mask image and the standard map mask image respectively, wherein the first matching point pair represents the matching point pair of the new positive binary mask image of each target mask image in the map to be matched and the standard map mask image, and the second matching point pair represents the matching point pair of the new debinarized mask image of each target mask image in the map to be matched and the standard map mask image.

[0094] In the embodiment of the present application, the modified new positive binary mask image and the new inverse binary mask image are respectively compared with the standard map mask image. Figure 1 The new positive binary mask image and the new inverse binary mask image are input into the image matching model (GlueSick), and the points and line segments of the new positive binary mask image and the new inverse binary mask image and the standard map mask image are extracted respectively, and the first matching point pair of the new positive binary mask image and the standard map mask image, as well as the second matching point pair of the new inverse binary mask image and the standard map mask image are determined.

[0095] Specifically, a common feature detector is used to extract points, lines and their descriptors and uniformly encode them into a wireframe. Then, a graph neural network is used to encode the appearance and spatial information of the input features, fully explore the association between points, and output matching feature descriptors with more information. Finally, two independent Softmax layers are used to match points and lines respectively to determine the matching correspondence. By processing two types of features together, the traditional paradigm that matching points and matching line segments between two views are regarded as two independent problems is broken. The correspondence between points and lines is jointly established in one framework, which makes the GlueStick method have a comprehensive improvement in the matching performance of points and lines. Based on the multi-target binary mask map to be matched, a loop traversal method is used to match it with the binary mask map of the standard map. Based on the majority matching strategy, the matching pair with the largest number of matching points is taken as the best match, and the most matching standard map is recommended for the map to be matched.

[0096] Step S8: determining a standard map that matches the map to be matched according to the first matching point pairs and the second matching point pairs of each target mask image.

[0097] In an embodiment of the present application, first matching point pairs and second matching point pairs are obtained respectively for the positive binarized mask images and the debinarized mask images of multiple standard map mask images and each target mask image, and the first matching point pairs and second matching point pairs obtained under the multiple target mask images are uniformly sorted to determine the standard map mask image with the most matching point pairs, and then the standard map that matches the map to be matched can be determined as the recommended map.

[0098] For example: Known target mask Figure 1 The new positive binary mask has 50 points, the new anti-binarized mask has 40 points, and the target mask Figure 2 The new positive binary mask has 60 points, the new anti-binarization mask has 35 points; the standard map mask Figure 1 There are 45 points, standard map mask Figure 2 There are 50 points.

[0099] Case 1: Mask the target Figure 1 Zhongxin positive binary mask map and standard map Figure 1 For association, it can correspond to 40 points;

[0100] Case 2: Mask the target Figure 1 Zhongxin debinarization mask map and standard map Figure 1 For association, it can correspond to 30 points;

[0101] Case 3: Mask the target Figure 1 Zhongxin positive binary mask map and standard map Figure 2 After correlation, 45 points can be matched;

[0102] Case 4: Mask the target Figure 1 Zhongxin debinarization mask map and standard map Figure 2 For association, it can correspond to 20 points;

[0103] Case 5: Mask the target Figure 2 Zhongxin positive binary mask map and standard map Figure 1 For association, it can correspond to 15 points;

[0104] Case 6: Mask the target Figure 2 Zhongxin debinarization mask map and standard map Figure 1 For association, it can correspond to 10 points;

[0105] Case 7: Mask the target Figure 2 Zhongxin positive binary mask map and standard map Figure 2For association, it can correspond to 5 points;

[0106] Case 8: Mask the target Figure 2 Zhongxin debinarization mask map and standard map Figure 2 For association, it can correspond to 25 points;

[0107] Then sort the matching points: Case 3 > Case 1 > Case 2 > Case 8 > Case 4 > Case 5 > Case 6 > Case 7. Therefore, it is determined that the scenario in Case 3 has the largest number of matching points and is the best match. The standard map mask Figure 2 The standard map is the standard map that best matches the map to be matched.

[0108] like Figure 4 and Figure 5 As shown in the figure, by performing positive and negative binarization processing on the target mask image, the complex information in the image is simplified into black and white colors, which makes the foreground and background clearly separated in the process of point and line segment feature comparison, enhances the contrast of the image, makes the part of interest more prominent, and can match with the standard map from different angles, thereby improving the accuracy of finding the standard map.

[0109] Furthermore, the map to be matched is grid-divided to generate a plurality of cue points; based on an image segmentation model, the image texture and block features of the map to be matched are extracted using the plurality of cue points; and according to the image texture and the block features, the map to be matched is segmented into a plurality of image segmentation mask images.

[0110] Specifically, the map to be matched is divided into multiple small grids according to certain rules (such as equal spacing or adaptive complexity based on map content), and prompt points are generated at the center or key position of each grid. These prompt points serve as reference points for subsequent image processing and feature extraction, which helps to more accurately locate and identify key elements in the map. The SAM image segmentation model is used to identify and distinguish different elements in the map (Beijing, Guangdong, Inner Mongolia, etc.), and the texture features of the grid area are extracted by calculating the gray level co-occurrence matrix (GLCM) and local binary pattern (LBP) of the pixels in the grid; combined with the position of the prompt point, a more detailed analysis is performed on the area around each grid or prompt point to extract block features such as shape, size, and color. These features can further describe the detailed attributes of the elements in the map. The extracted image texture features and block features are then fused to form a comprehensive feature representation of each grid or prompt point area. Each grid or prompt point area is classified according to the fused features. After that, a category label is assigned to each pixel of the map to be matched according to the classification results, and multiple image segmentation mask images are generated.

[0111] Furthermore, the area of ​​each of the image segmentation mask images is determined; the image segmentation mask images are sorted according to the area; and the image segmentation mask images that are larger than a threshold are selected as target mask images.

[0112] Specifically, for each image segmentation mask map, the area is determined, and then a sorting algorithm is used to sort the image segmentation mask maps according to the area size, and a plurality of mask maps with larger areas are selected using a threshold and determined as target mask maps.

[0113] Furthermore, if the number of pixels of width and height of the standard map mask are different, the number of pixels of height and width of the standard map mask are obtained according to the standard map mask; the minimum number of pixels between the number of pixels of height and the number of pixels of width in the standard map mask is determined; the number of pixels of each target mask is obtained respectively; wherein the number of pixels of the target mask includes the number of pixels of height and the number of pixels of width of the target mask; according to the minimum number of pixels in the standard map mask, a first proportional adjustment benchmark is determined, the first proportional adjustment benchmark being the height or width of the standard map mask; according to the first proportional adjustment benchmark and the number of pixels of each target mask, a first adjustment ratio is determined; according to the first adjustment ratio, the number of pixels of width and the number of pixels of height of the positive binary mask and the debinarized mask of each target mask are adjusted respectively. Enlarge or reduce the image by the same proportion to generate a new positive binary mask image and a new inverse binary mask image; if the width pixel number and the height pixel number of the standard map mask image are the same, obtain the height pixel number and the width pixel number of the target mask image according to the target mask image; determine the minimum pixel between the height pixel number and the width pixel number in the target mask image; determine a second proportion adjustment reference according to the minimum pixel number in the target mask image, the second proportion adjustment reference being the height or width in the target mask image; determine a second adjustment ratio according to the second proportion adjustment reference and the number of pixels of each standard map mask image; according to the second adjustment ratio, enlarge or reduce the width pixel number and the height pixel number of the positive binary mask image and the inverse binary mask image of each target mask image by the same proportion to generate a new positive binary mask image and a new inverse binary mask image.

[0114] Specifically, in case 1, if the width pixel number and height pixel number of the standard map mask are different, first, the height pixel number and width pixel number are obtained from the standard map mask. Among the obtained height pixel number and width pixel number, find the smaller value as the minimum pixel number, and this value will be used as the basis for the subsequent adjustment ratio. Then obtain the height pixel number and width pixel number of each target mask, and use the minimum pixel number as the basis to calculate the proportional adjustment coefficient of each target mask relative to the standard map mask. Secondly, according to the calculated adjustment ratio, the width and height of the positive binary mask and the debinarized mask of each target mask are enlarged or reduced in the same proportion, so that the size of all target mask images maintains a certain proportional relationship with the size of the standard map mask, so that the subsequent matching point pair association is more accurate. Finally, after the adjustment of the above steps, each target mask will generate a new positive binary mask and a new debinarized mask, and these new mask will have a similar proportional relationship with the standard map mask. Case 2: If the width pixel number and height pixel number of the standard map mask are the same, compare the height pixel number and width pixel number in the target mask pixel number to find the minimum pixel number, and then calculate the adjustment ratio to generate a new positive binary mask and a new inverse binary mask.

[0115] Furthermore, the image matching model includes a feature detector and a graph neural network model, and the feature detector is used to extract point and line segment features of the new positive binarization mask image, the new debinarization mask image and the standard map mask image; and based on the graph neural network model, the point and line segment features of the new positive binarization mask image and the new debinarization mask image are matched with the point and line segment features of the standard map mask image, and the first matching point pair and the second matching point pair of the new positive binarization mask image and the new debinarization mask image and the standard map mask image are output respectively.

[0116] Specifically, the feature detector is used to first extract the feature points and line segment features of the new positive binary pickled film image and the new reverse binary pickled film image, and then the point and line segment features of the standard map pickled film image in the preset standard map library are respectively extracted, and then the graph neural network model is used to match the new positive and reverse binary point and line segment features with the point and line segment features in the standard map library, thereby obtaining the first matching point pair of the new positive binary pickled film image and each standard map pickled film image in the preset standard map library, and the second matching point pair of the new reverse binary pickled film image and each standard map pickled film image in the preset standard map library. Through this method, each standard map in the preset standard map library can be matched with the map to be matched, thereby reducing the inaccuracy of the matching result.

[0117] Furthermore, the preset standard map library includes multiple standard maps; each of the standard maps is binarized to obtain multiple standard map mask images; for each standard map mask image, the size of the current standard map mask image is determined; according to the size of the current standard map mask image, the sizes of the positive binarization mask image and the debinarization mask image are updated to obtain the new positive binarization mask image and the new debinarization mask image; according to the current standard map mask image and the new positive binarization mask image, the first matching point pair is determined; and according to the current standard map mask image and the new debinarization mask image, the second matching point pair is determined.

[0118] Specifically, binarization is performed on each standard map in each preset standard map library, and a standard map pick-up map is obtained. The size of each standard map pick-up map, as well as the size of the positive binary pick-up map and the inverse binary pick-up map are measured. Before comparing the map to be tested with the standard map, the size of the positive binary pick-up map is modified based on the size of the standard map pick-up map; then, the point and line segment features on the boundary of the standard map pick-up map and the new positive binary pick-up map are extracted respectively, and the point and line segment features on the standard map pick-up map are paired with the point and line segment features on the new positive binary pick-up map to obtain multiple matching point pairs (first matching point pairs). Then, based on the size of the standard map pick-up map, the size of the inverse binary pick-up map is modified; the point and line segment features on the boundary of the standard map pick-up map and the new inverse binary pick-up map are extracted respectively, and the point and line segment features on the standard map pick-up map are paired with the point and line segment features on the new inverse binary pick-up map to obtain multiple matching point pairs (second matching point pairs).

[0119] Furthermore, the first matching point pairs and the second matching point pairs of each of the standard map mask images and each of the target mask images are sorted respectively, and the standard map having the most matching point pairs when matching with the target mask image is output.

[0120] Specifically, there are multiple standard map pickled film images in the preset standard map library, and each standard map pickled film image is respectively matched with the new positive binary pickled film image and the new reverse binary pickled film image in the map to be matched to obtain the first matching point pair and the second matching point pair. It should be noted that different first matching point pairs and second matching point pairs are obtained when different standard map pickled film images are matched with the new positive binary pickled film image and the new reverse binary pickled film image. These different first matching point pairs and second matching point pairs are then sorted, and the one with the most matching point pairs is selected, and its standard map is output. This standard map is the map that is most similar to the map to be matched.

[0121] Further, a distributed computing framework is used in step S5 and steps S6 to S8.

[0122] Specifically, after obtaining the pickled film images of each standard map in the preset standard map library, distributed computing is used to traverse the standard maps in the preset standard map library, and each standard map is binarized in parallel to determine the size of the binarized standard map pickled film image, and the size of the positive binarized pickled film image and the de-binarized pickled film image is modified according to the size, and the point and line segment features of the standard map pickled film image, the new positive binarized pickled film image, and the new de-binarized pickled film image are extracted, and the point and line segment features of the new positive binarized pickled film image and the new de-binarized pickled film image are matched with the point and line segment features of the standard map pickled film image, respectively, to obtain different first matching point pairs and second matching point pairs, and then these matching point pairs are sorted to obtain the standard map.

[0123] For example: The preset standard map library contains standard maps Figure 1 , standard Figure 2 , standard Figure 3 , respectively, and perform binarization to obtain the standard map pickling film Figure 1 , Standard map pickling film Figure 2 , Standard map pickling film Figure 3 In parallel processing, multiple nodes simultaneously perform multiple target new positive and negative binary mask map matching operations with the standard map. Figure 2 The binary mask map is stored in advance to reduce the multiple segmentation and masking processes that need to be matched with the standard map, thereby improving work efficiency. Figure 1 Then the point and line segment features of the new positive binary pickled film image and the new inverse binary pickled film image are compared with the standard map pickled film Figure 1 Node 2: Modify the size of the positive binary pickled film image and the inverse binary pickled film image to make them consistent with the standard map pickled film image. Figure 2 Consistent, then get the standard map pickling film Figure 2 , respectively compare the point and line segment features of the new positive binary pickled film image and the new inverse binary pickled film image with the standard map pickled film Figure 2 Node 3: Modify the size of the positive binary pickled film image and the inverse binary pickled film image to make them consistent with the standard map pickled film image. Figure 3 Consistent, then get the standard map pickling film Figure 3 , respectively compare the point and line segment features of the new positive binary pickled film image and the new inverse binary pickled film image with the standard map pickled film Figure 3 The point and line segment features are matched to obtain the first matching point pair and the second matching point pair. The above matching is performed on the positive and negative binary mask images of multiple targets in a cyclic matching operation, and then the first matching point pair and the second matching point pair of each channel are uniformly sorted to obtain the standard map that is most similar to the map to be matched.

[0124] The present application embodiment provides a map matching device, such as Figure 6 As shown, including:

[0125] A first acquisition module 101 is used to acquire a map to be matched;

[0126] A first determination module 102 is used to input the to-be-matched map into an image segmentation model to obtain a plurality of image segmentation mask maps;

[0127] A second determination module 103 is used to determine a plurality of target mask images greater than a threshold from the plurality of image segmentation mask images;

[0128] The third determination module 104 is used to perform positive and negative binarization processing on each of the target mask images to obtain a positive binarization mask image and a negative binarization mask image;

[0129] The second acquisition module 105 is used to acquire a standard map mask image and a size of the standard map mask image for each standard map in the preset standard map library;

[0130] The fourth determination module 106 is used to generate a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size; wherein the size represents the number of pixels in height and the number of pixels in width; and

[0131] A fifth determination module 107 is used to match the generated new positive binary mask image and the new debinarized mask image with the standard map mask image respectively based on the image matching model, and output the first matching point pair and the second matching point pair of each target mask image and the standard map mask image respectively, wherein the first matching point pair represents the matching point pair of the new positive binary mask image of each target mask image in the map to be matched and the standard map mask image, and the second matching point pair represents the matching point pair of the new debinarized mask image of each target mask image in the map to be matched and the standard map mask image; and

[0132] The sixth determination module 108 is configured to determine a standard map that matches the map to be matched according to the first matching point pairs and the second matching point pairs of each of the target mask images.

[0133] An embodiment of the present application provides a storage medium for storing computer-executable instructions, wherein the computer-executable instructions implement the steps of the map matching method described in any one of the above embodiments when executed.

[0134] It should be noted that the embodiments of the storage medium in this specification and the embodiments of the blockchain-based service provision method in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the aforementioned corresponding implementation of the blockchain-based service provision method, and the repeated parts will not be repeated.

[0135] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0136] In the 1930s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0137] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0138] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0139] For the convenience of description, the above devices are described in terms of functions and are divided into various units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0140] It should be understood by those skilled in the art that one or more embodiments of this specification may be provided as a method, system or computer program product. Therefore, one or more embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0142] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0143] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0144] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0145] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0146] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0147] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0148] One or more embodiments of the present specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0149] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0150] The above description is only an embodiment of this document and is not intended to limit this document. For those skilled in the art, this document may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this document should be included in the scope of the claims of this document.

Claims

1. A map matching method, characterized in that include: S1. Obtain the map to be matched; S2, inputting the map to be matched into an image segmentation model to obtain a plurality of image segmentation mask maps; S3, determining a plurality of target mask images greater than a threshold value from the plurality of image segmentation mask images; S4, performing positive and negative binarization processing on each of the target mask images to obtain a positive binarization mask image and a negative binarization mask image; S5. For each standard map in the preset standard map library, obtain a standard map mask image and a size of the standard map mask image; S6. For each standard map in the preset standard map library, generate a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size; wherein the size represents the number of pixels in height and the number of pixels in width; S7. Based on the image matching model, the generated new positive binary mask image and the new debinarized mask image are matched with the standard map mask image respectively, and the first matching point pair and the second matching point pair of each target mask image and the standard map mask image are output respectively, wherein the first matching point pair represents the matching point pair of the new positive binary mask image of each target mask image in the map to be matched and the standard map mask image, and the second matching point pair represents the matching point pair of the new debinarized mask image of each target mask image in the map to be matched and the standard map mask image; and S8. Determine a standard map that matches the map to be matched according to the first matching point pairs and the second matching point pairs of each of the target mask images.

2. The method according to claim 1, characterized in that The step of inputting the to-be-matched map into an image segmentation model to obtain a plurality of image segmentation mask maps comprises: Dividing the to-be-matched map into grids to generate a plurality of prompt points; Based on the image segmentation model, using multiple prompt points, extracting the image texture and tile features of the map to be matched; and According to the image texture and the image block features, the map to be matched is segmented into a plurality of image segmentation mask images.

3. The method according to claim 1, characterized in that The determining of a plurality of target mask images greater than a threshold value according to the plurality of image segmentation mask images comprises: Determining the area of ​​each of the image segmentation mask images; sorting the image segmentation mask images according to the areas; and The image segmentation mask image that is greater than a threshold is selected as the target mask image.

4. The method according to claim 1, characterized in that: For each standard map in the preset standard map library, generating a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size includes: If the width and height of the standard map mask are different, According to the standard map mask, the number of pixels of the height of the standard map mask and the number of pixels of the width of the standard map mask are obtained; Determine the minimum number of pixels between the number of pixels of the height and the number of pixels of the width in the standard map mask image; The number of pixels of each target mask image is obtained respectively; wherein the number of pixels of the target mask image includes the number of pixels of the height of the target mask image and the number of pixels of the width of the target mask image; Determining a first scale adjustment reference according to the minimum number of pixels in the standard map mask image, wherein the first scale adjustment reference is the height or width of the standard map mask image; Determining a first adjustment ratio according to the first ratio adjustment reference and the number of pixels of each of the target mask images; According to the first adjustment ratio, the number of pixels in width and the number of pixels in height of the positive binary mask image and the debinarized mask image of each of the target mask images are respectively enlarged or reduced in the same ratio to generate a new positive binary mask image and a new debinarized mask image; If the width and height of the standard map mask are the same, According to the target mask image, the number of pixels of the height of the target mask image and the number of pixels of the width of the target mask image are obtained; Determine the minimum pixel between the number of height pixels and the number of width pixels in the target mask image; Determining a second ratio adjustment reference according to the minimum number of pixels in the target mask image, where the second ratio adjustment reference is the height or width in the target mask image; Determining a second adjustment ratio according to the second ratio adjustment reference and the number of pixels of each standard map mask image; According to the second adjustment ratio, the width pixel number and the height pixel number of the positive binary mask image and the debinarized mask image of each of the target mask images are respectively enlarged or reduced in the same proportion to generate a new positive binary mask image and a new debinarized mask image.

5. The method according to claim 1, characterized in that The method of matching the generated new positive binary mask image and the generated new inverse binary mask image with the standard map mask image respectively based on the image matching model to output the first matching point pair and the second matching point pair of each target mask image and the standard map mask image respectively includes: the image matching model includes a feature detector and a graph neural network model, and Using a feature detector, extracting point and line segment features of the new positive binarization mask image, the new debinarization mask image, and the standard map mask image; and Based on the graph neural network model, the point and line segment features of the new positive binary mask image and the new debinarized mask image are matched with the point and line segment features of the standard map mask image, and the first matching point pair and the second matching point pair of the new positive binary mask image and the new debinarized mask image and the standard map mask image are output respectively.

6. The method according to claim 1, characterized in that The preset standard map library includes a plurality of standard maps; binarization is performed on each of the standard maps to obtain a plurality of standard map mask images; and Based on the image matching model, the generated new positive binary mask image and the new inverse binary mask image are matched with the standard map mask image respectively, and the first matching point pair and the second matching point pair of each target mask image and the standard map mask image are output respectively, including: For each standard map mask, determining the size of the current standard map mask; According to the size of the current standard map mask image, the sizes of the positive binary mask image and the debinarized mask image are updated to obtain the new positive binary mask image and the new debinarized mask image; Determining the first matching point pair according to the current standard map mask image and the new positive binary mask image; and The second matching point pair is determined according to the current standard map mask image and the new debinarized mask image.

7. The method according to claim 6, characterized in that The determining of a standard map matching the map to be matched according to the first matching point pairs and the second matching point pairs of each target mask image comprises: The first matching point pairs and the second matching point pairs of each of the standard map mask images and each of the target mask images are sorted respectively, and a standard map having the most matching point pairs when matching with the target mask image is output.

8. The method according to claim 1, wherein step S5 and steps S6 to S8 are executed using a distributed computing framework.

9. A map matching device, characterized in that include: A first acquisition module is used to acquire a map to be matched; A first determination module is used to input the map to be matched into an image segmentation model to obtain a plurality of image segmentation mask maps; A second determination module is used to determine a plurality of target mask images greater than a threshold from the plurality of image segmentation mask images; A third determination module is used to perform positive and negative binarization processing on each of the target mask images to obtain a positive binarization mask image and a negative binarization mask image; A second acquisition module is used to acquire a standard map mask image and a size of the standard map mask image for each standard map in a preset standard map library; A fourth determination module is used to generate a new positive binary mask image and a new debinarized mask image of each target mask image according to the standard map mask image and the size; wherein the size represents the number of pixels in height and the number of pixels in width; and a fifth determination module, for matching the generated new positive binary mask image and the new debinarized mask image with the standard map mask image respectively based on the image matching model, and outputting first matching point pairs and second matching point pairs of each target mask image and the standard map mask image respectively, wherein the first matching point pairs represent matching point pairs of the new positive binary mask image of each target mask image in the map to be matched and the standard map mask image, and the second matching point pairs represent matching point pairs of the new debinarized mask image of each target mask image in the map to be matched and the standard map mask image; and The sixth determination module is used to determine a standard map that matches the map to be matched according to the first matching point pairs and the second matching point pairs of each target mask image.

10. A storage medium for storing computer executable instructions, characterized in that: When the computer executable instructions are executed, the steps of the map matching method according to any one of claims 1 to 7 are implemented.

Images