An electronic map production method, production device, equipment and medium

Abstract

The application provides a method and device for making an electronic map, equipment and a medium. The method comprises: obtaining road data detected by a road detection tool in a target area, wherein the road data comprises a plurality of road images and a plurality of road positioning data; for each road image, identifying road indication information of a road sign in the road image, and determining an actual position of the road sign according to road positioning data matched with the road image; generating an initial road network of the target area according to a plurality of the road positioning data; and adding the road indication information of each road sign to the initial road network according to the actual position of each road sign, to obtain an electronic map of the target area.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a method and device for making an electronic map, and a medium. BACKGROUND

[0002] Roads are the foundation of social development, so with the progress of society, more and more roads are being built, including expressways, provincial roads, national roads, county roads, and rural roads. In order to prevent people from getting lost and facilitate travel, electronic maps have been developed. Based on the actual location of each road in a real-world scenario, digital processing can be performed to obtain an electronic map. Traditional methods of making electronic maps mainly rely on map surveying of each region, but there are many problems such as data collection, verification, error diagnosis, etc., which make it difficult to make electronic maps. SUMMARY

[0003] Therefore, the present application aims to provide a method and device for making an electronic map, and a medium, to solve the problem of difficulty in making electronic maps in the prior art.

[0004] In a first aspect, the present application provides a method for making an electronic map, comprising:

[0005] obtaining road data detected by a road detection tool in a target area, the road data comprising a plurality of road images and a plurality of road positioning data;

[0006] for each road image, identifying road indication information of a road sign in the road image, and determining an actual position of the road sign according to road positioning data matched with the road image;

[0007] generating an initial road network of the target area according to the plurality of road positioning data;

[0008] adding the road indication information of each road sign to the initial road network according to the actual position of each road sign, to obtain an electronic map of the target area.

[0009] Optionally, for each road image, identifying road indication information of a road sign in the road image comprises:

[0010] for each road image, determining a to-be-identified region in the road image according to the position of a horizontal line in the road image;

[0011] for each road image, determining an effective image region containing a road sign in the to-be-identified region using an image recognition model;

[0012] For each road image, the character recognition model is used to determine road indication information of the road sign in the effective image region.

[0013] Optionally, for each road image, the character recognition model is used to determine road indication information of the road sign in the effective image region, including:

[0014] For each road image, the effective image region is morphologically corrected to obtain a front view of the road sign.

[0015] For each road image, the character recognition model is used to determine road indication information of the road sign in the front view of the road sign.

[0016] Optionally, for each road image, the effective image region is morphologically corrected to obtain a front view of the road sign, including:

[0017] For each road image, an image cutting algorithm is used to determine a road sign region and a background region in the effective image region.

[0018] For each road image, according to the position of the boundary of the road sign region in the effective image region, the vertex positions of the four vertices of the road sign in the effective image region are determined.

[0019] For each road image, the road sign region is morphologically corrected by using the vertex positions of the four vertices of the road sign to obtain a front view of the road sign.

[0020] Optionally, for each road image, the actual position of the road sign is determined according to the road positioning data matched with the road image, including:

[0021] For each road image, the actual position of the road sign is determined according to the road positioning data matched with the road image and the road positioning data matched with the historical road image, and the difference between the shooting time of the historical road image and the shooting time of the road image meets a preset requirement.

[0022] In a second aspect, an embodiment of the present application provides a device for making an electronic map, including:

[0023] The acquisition module is configured to acquire road data detected by a road detection tool in a target region, the road data including a plurality of road images and a plurality of road positioning data.

[0024] The determining module is configured to, for each road image, identify road indication information of a road sign in the road image, and determine an actual position of the road sign according to road positioning data matched with the road image.

[0025] The generating module is configured to generate an initial road network of the target area according to the plurality of road positioning data.

[0026] The adding module is configured to add the road indication information of each road sign to the initial road network according to the actual position of each road sign, to obtain the electronic map of the target area.

[0027] Optionally, the determining module comprises:

[0028] The first determining unit is configured to, for each road image, determine a to-be-identified region in the road image according to a horizontal line position in the road image.

[0029] The second determining unit is configured to, for each road image, determine an effective image region containing a road sign in the to-be-identified region by using an image recognition model.

[0030] The third determining unit is configured to, for each road image, determine road indication information of a road sign in the effective image region by using a character recognition model.

[0031] Optionally, the third determining unit comprises:

[0032] The first determining subunit is configured to, for each road image, perform morphological correction on the effective image region to obtain a main view of the road sign.

[0033] The second determining subunit is configured to, for each road image, determine road indication information of a road sign in the main view of the road sign by using a character recognition model.

[0034] In a third aspect, an embodiment of the present application provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements steps of the above method when executing the computer program.

[0035] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program runs on a processor to execute steps of the above method.

[0036] The method for manufacturing an electronic map provided by the embodiment of the present application firstly acquires road data detected by a road detection tool in a target area, wherein the road data comprises a plurality of road images and a plurality of road positioning data; secondly, for each road image, road indication information of a road sign in the road image is identified, and an actual position of the road sign is determined according to road positioning data matched with the road image; thirdly, an initial road network of the target area is generated according to a plurality of the road positioning data; and finally, the road indication information of each road sign is added to the initial road network according to the actual position of each road sign, so as to obtain an electronic map of the target area.

[0037] In some embodiments, the present application directly acquires road images and road positioning data through a road detection tool, then directly identifies road indication information of a road sign from the road images, and calculates an actual position of the road sign based on the road positioning data, so that the road indication information of the road sign can be directly mapped to an initial road network generated based on the road positioning data, and then an electronic map is generated, without the need of map surveying and data format integration, thereby improving the convenience of manufacturing the electronic map.

[0038] In order to make the above objectives, characteristics and advantages of the present application more apparent and comprehensible, the following preferred embodiments are specifically described below, and the accompanying drawings are referred to for detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be considered as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0040] Figure 1 A flowchart of a method for manufacturing an electronic map provided by the embodiment of the present application;

[0041] Figure 2 A schematic diagram of a road sign provided by the embodiment of the present application;

[0042] Figure 3 A structural schematic diagram of a manufacturing device for an electronic map provided by the embodiment of the present application;

[0043] Figure 4 A structural schematic diagram of a computer device provided by the embodiment of the present application;

[0044] Figure 5 A schematic diagram of a road image provided by the embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will be combined with the accompanying drawings for the embodiments of the present application to make a clear and complete description of the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application and are not all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0046] The electronic map in the prior art is basically generated based on map surveying technology. In the process of making the electronic map, a large amount of data needs to be measured manually, and the measured data needs to be checked multiple times. Even so, there are still error data, which leads to the problem that the made electronic map may be inaccurate. In addition, when making the electronic map, the data needs to be sorted into the same data format. Due to the large amount of data and the large difference in the types of data collection equipment, the data sorting is relatively cumbersome.

[0047] Based on the above defects, the embodiments of the present application provide a method for making an electronic map, as shown in the following Figure 1 The method comprises the following steps:

[0048] S101, acquiring road data detected by a road detection tool in a target area, wherein the road data comprises a plurality of road images and road positioning data;

[0049] S102, for each road image, identifying road indication information of a road indication board in the road image, and determining an actual position of the road indication board according to the road positioning data corresponding to the road image;

[0050] S103, generating an initial road network of the target area according to the shooting positions of the plurality of road images;

[0051] S104, adding the road indication information of each road indication board to the initial road network according to the actual position of each road indication board, to obtain an electronic map of the target area.

[0052] In the step S101, the road detection tool is used to obtain the road data in the target region. The road detection tool can be a vehicle equipped with a camera, and a GPS (Global Positioning System) is also installed on the road detection tool. The road positioning data is the data obtained by the GPS, i.e., the geographic position information, and the road positioning data also carries the positioning time for determining the time when each geographic position information is obtained. The road image is obtained by the camera, and the road image carries the shooting time. The road image includes the scene information of the street captured by the camera, and the scene in the image can include the following objects: buildings, moving vehicles (vehicles), people, traffic lights, road signs, etc. The road detection tool captures the road image by the camera and obtains the current position by the GPS. Since the camera and the vehicle are rigidly connected, the position of the road detection tool when the road image is captured by the camera can be determined based on the shooting time of the road image captured by the camera.

[0053] In the step S102, the road sign is a sign set on the roadside in the real scene, which can play a prompting role for people, such as Figure 2 As shown (the road sign shows the directions of AAA West Road, BBB Boulevard, CCC Street, and DDD Bridge). The road indication information of the road sign represents the road information of the road where the road sign is located, including the road name, the direction indication (such as which direction is the north), etc.

[0054] In the implementation, the road sign set on the roadside can be captured in the road image, and the road indication information is recorded in the road sign. The text information in the image can be recognized by a text recognition model. If the road sign is captured in the road image, the road indication information of the road sign can be recognized from the road image. According to the shooting time of the road image and the positioning time of the road positioning data, the road positioning data obtained by the GPS when the road image is captured can be determined, i.e., the road positioning data determined by the shooting time is the shooting position when the road image is captured by the camera (equivalent to the geographic position of the road detection tool when the road image is captured by the camera). When the road detection tool travels on a road, the camera will continuously and uninterruptedly capture, so multiple road images will capture the scene information at different positions in the same road, and the road sign is set on the roadside, so it is possible that the same road sign is captured in multiple road images. Based on the shooting positions of the multiple road images capturing the same road sign, the actual position of the road sign in the real scene can be calculated.

[0055] In the step S103, the road detection tool splices the plurality of road positioning data acquired by the road detection tool based on the road positioning data acquired by moving in the target area, to obtain the electronic data of each road in the target area, that is, the initial road network, and the initial road network records the actual geographical position of each road in the target area.

[0056] In the step S104, the actual position of each road sign is determined in the step S102, and the initial road network is generated based on the actual position of each road in the target area. The mapping position of the road sign in the initial road network can be determined based on the actual position of each road sign, and then the indication information of the road sign is added to the mapping position of the initial road network, so as to obtain the electronic map matched with the actual road scene.

[0057] The method for making the electronic map provided in the application directly acquires the road image and the road positioning data by the road detection tool, directly identifies the road indication information of the road sign from the road image, and calculates the actual position of the road sign based on the road positioning data. Thus, the road indication information of the road sign can be directly mapped to the initial road network generated based on the road positioning data, and the electronic map is generated, without the need of map surveying and data format integration, so as to improve the convenience and accuracy of making the electronic map.

[0058] The road map is a relatively large image, and not only the road sign but also other vehicles, buildings, etc. are photographed in the image. In order to improve the identification efficiency of the road indication information, the following processing can be performed, that is, the step S102 comprises:

[0059] In the step 1021, for each road image, the horizontal line position in the road image is determined, and the to-be-identified region in the road image is determined according to the horizontal line position.

[0060] In the step 1022, for each road image, the effective image region containing the road sign in the to-be-identified region is determined by using the image recognition model.

[0061] In the step 1023, for each road image, the road indication information of the road sign in the effective image region is determined by using the character recognition model.

[0062] In the step 1021, the road sign is arranged at a position far from the ground, and thus the road image can be divided by the horizontal line in the road image, and the region above the horizontal line is determined as the to-be-identified region, that is, only the part above the horizontal line in the road image is identified, as shown in FIG. 2. Figure 5 As shown in FIG. 2, the road sign is arranged above the horizontal line in the road image.

[0063] In step 1022, the image recognition model is used to determine whether there is a road sign in the to-be-recognized region. The road sign is generally rectangular in shape and has a relatively prominent bright color in the non-character region, for example, the bright color can be blue. The image recognition model can be trained through a large number of training samples. In the specific training process, the region where the road sign is located in the training sample is marked, and the image recognition model is trained by marking the road sign to learn the features of the road sign.

[0064] In order to improve the recognition accuracy of the image recognition model, the size of the image input to the image recognition model is controlled. If the image input to the image recognition model is too large, it may interfere with the image recognition because there is too much content in the image. If the image input to the image recognition model is too small, it may not be able to completely contain the road sign in the image, and thus accurate recognition of the road sign cannot be achieved. Therefore, for each road image, the to-be-recognized region is divided using a preset image size (that is, the image size of the training sample of the image recognition model, which is artificially set). Each divided image region is input to the image recognition model, and then an effective image region containing the road sign is determined.

[0065] In step 1023, the character recognition model can be an OCR (optical character recognition) model, which is used to recognize the characters in the image. In the character recognition model, a resnet network (deep residual network) is used, and a BiLSTM (Bi-directional Long Short-Term Memory) model is used as the language model. A decoder based on bahdanau attention is used in the decoding module, and two LSTM (long short-term memory) networks are used for left-to-right and right-to-left bidirectional decoding.

[0066] In a specific implementation, the effective image region containing the road sign can be determined from the to-be-recognized region in step 1022, and the image size is further reduced, and then the character recognition model is used to accurately recognize the road indication information from the effective image region.

[0067] Since the image captured by the camera of the road detection tool is deformed, in order to ensure the accuracy of the road indication information recognition, the deformed image is corrected, that is, step 1023 includes:

[0068] In step 10231, for each road image, the effective image region is deformed and corrected to obtain a main view of the road sign.

[0069] Step 10232: For each road image, use a character recognition model to determine the road instruction information of the road sign in the main view of the road sign.

[0070] In step 10231 above, the front view is the image obtained from the view directly in front of the road sign.

[0071] In practice, deformation correction includes the following steps:

[0072] Step 11: For each road image, use an image segmentation algorithm to divide the pixels in the effective image area into a road sign area and a background area;

[0073] Step 12: For each road image, determine the vertex positions of the four vertices of the road sign in the effective image area based on the position of the boundary of the road sign area in the effective image area;

[0074] Step 13: For each road image, use the vertex positions of the four vertices of the road sign to perform deformation correction on the road sign area to obtain the front view of the road sign.

[0075] In step 11 above, the image segmentation algorithm is used to distinguish the pixels in the effective image region into pixels in the road sign region and pixels in the background region. The image segmentation algorithm can be a graph cut algorithm. In the image segmentation algorithm, the center point of the effective image region is used as the foreground seed point, and the four boundary vertices of the effective image region are used as the background seed points. Then, the energy function of the image is used to determine the foreground region, which is the road sign region, in the effective image region.

[0076] The energy function of the image is: E(L)=aR(L)+B(L);

[0077] L is the set of pixels of the secant line, L = {l1, l2, ..., l...} i}, l i The value of is 0 or 1, the value of s (foreground) is 1, the value of t (background) is 0, E(L) is the energy loss function after segmentation in L, R(L) is the region term, B(L) is the boundary term, and a is the weight factor.

[0078] In step 12, after determining the road sign area, in order to more accurately determine the vertex positions of the four vertices of the road sign, the position of the boundary of the road sign area can be determined first. The specific steps for determining the position of the boundary of the road sign area include: performing gray processing on the effective image area, then performing binary processing on the road sign area and the background area in the gray-processed effective image area, that is, setting the pixel points of the road sign area to 1 and setting the pixel points of the background area to 0, and then using hough transformation to extract the four boundaries of the road sign area in the binary image. Then, the four vertices intersected by the four boundaries are calculated by using the calculation method of line intersection.

[0079] In step 13, after determining the vertex positions of the four vertices of the road sign, the four edges of the road sign in the effective image area can be corrected to a regular rectangle by using inverse projection transformation. Inverse projection transformation is a transformation method for inversely mapping the projection transformation existing in the process of collecting the actual scene image by the camera using the internal and external parameters of the camera. The application scene image collected by the camera is the reflection mapping of each object in the world coordinate system to the camera plane, and the inverse projection transformation is the inverse mapping of the image collected by the camera to obtain the actual physical coordinates of the object in the image in the world coordinate system. Then, the difference transformation is used to map all the pixel points contained in the quadrilateral of the road sign in the effective image area to the corrected rectangle, so as to realize the projection of the deformed road sign area into the main view of the road sign.

[0080] During driving, the road detection tool will take pictures of the road sign at different positions. In order to accurately calculate the position of the road sign, multiple road images can be used for calculation, that is, step S102 includes:

[0081] In step 1024, for each road image, the actual position of the road sign is determined according to the road positioning data matched with the road image and the road positioning data matched with the historical road image. The difference between the shooting time of the historical road image and the shooting time of the road image meets the preset requirement.

[0082] In step 1024, in the road image, the actual distance (that is, the distance in the real scene) between the camera and the road sign can be calculated according to the position of the road sign in the image, and then the actual position of the pixel point in the image can be further calculated. The specific pixel equation is:

[0083] p = (x + f · a, y + f · b, z + f · g);

[0084] Wherein, (x, y, z) is the road positioning data matched with the road image and (a, b, g) is the Euler angle detected by the inertial navigation in the road detection tool at the time of shooting the road image, P is the position information of the pixel point in the actual scene, and f is the distance from the road sign to the road detection tool.

[0085] Two pictures can confirm the position of the same road name and direction signboard, and can also check the content in both directions. The specific calculation formula is:

[0086]

[0087] Wherein, (x1, y1, z1) is the road positioning data matched with the road image, (a1, b1, g1) is the Euler angle detected by the inertial navigation in the road detection tool at the time of shooting the road image, f i is the distance between the road sign and (x1, y1, z1), (x2, y2, z2) is the road positioning data matched with the historical road image, and (a2, b2, g2) is the Euler angle detected by the inertial navigation in the road detection tool at the time of shooting the historical road image, f j is the distance from the road sign to (x1, y1, z1), and (x3, y3, z3) is the actual position of the road sign.

[0088] Based on the electronic map production method mentioned in the above embodiment, the embodiment of the application further provides an electronic map production device, as shown in Figure 3 The device comprises:

[0089] The acquisition module 301 is configured to acquire road data detected by the road detection tool in the target area, wherein the road data comprises a plurality of road images and a plurality of road positioning data.

[0090] The determination module 302 is configured to, for each road image, identify road indication information of a road sign in the road image, and determine an actual position of the road sign according to road positioning data matched with the road image.

[0091] The generation module 303 is configured to generate an initial road network of the target area according to a plurality of the road positioning data.

[0092] The addition module 304 is configured to add road indication information of each road sign to the initial road network according to the actual position of each road sign, so as to obtain an electronic map of the target area.

[0093] Optionally, the determination module comprises:

[0094] The first determining unit is configured to determine, for each road image, a region to be recognized in the road image according to a horizontal line position in the road image.

[0095] The second determining unit is configured to determine, for each road image, an effective image region containing a road sign in the region to be recognized by using an image recognition model.

[0096] The third determining unit is configured to determine, for each road image, road sign information of the road sign in the effective image region by using a character recognition model.

[0097] Optionally, the third determining unit comprises:

[0098] The first determining sub-unit is configured to perform morphological correction on the effective image region to obtain a front view of the road sign for each road image.

[0099] The second determining sub-unit is configured to determine, for each road image, road sign information of the road sign in the front view of the road sign by using a character recognition model.

[0100] Optionally, the first determining sub-unit comprises:

[0101] The third determining sub-unit is configured to determine, for each road image, a road sign region and a background region in the effective image region by using an image cutting algorithm.

[0102] The fourth determining sub-unit is configured to determine, for each road image, vertex positions of four vertices of the road sign in the effective image region according to a position of a boundary of the road sign region in the effective image region.

[0103] The fifth determining sub-unit is configured to perform morphological correction on the road sign region by using the vertex positions of the four vertices of the road sign to obtain a front view of the road sign for each road image.

[0104] Optionally, the determining module comprises:

[0105] The fourth determining unit is configured to determine, for each road image, an actual position of the road sign according to road positioning data matched with the road image and road positioning data matched with a historical road image, wherein a difference between a shooting time of the historical road image and a shooting time of the road image meets a preset requirement.

[0106] Corresponding to the electronic map making method in Figure 1 The present application also provides a computer device 400, as shown in Figure 4As shown, the device includes a memory 401, a processor 402, and a computer program stored in the memory 401 and executable on the processor 402, wherein the processor 402 executes the computer program to implement the method for creating the electronic map.

[0107] Specifically, the memory 401 and processor 402 can be general-purpose memory and processor, without any specific limitations. When the processor 402 runs the computer program stored in the memory 401, it can execute the above-mentioned method for making electronic maps, which solves the problem that the making of electronic maps is relatively difficult in the prior art.

[0108] Corresponding to Figure 1 In addition to the method for creating electronic maps, this application also provides a computer-readable storage medium storing a computer program, which is executed by a processor to perform the steps of the above-described method for creating electronic maps.

[0109] Specifically, the storage medium can be a general-purpose storage medium, such as a portable disk or hard disk. When the computer program on the storage medium is run, it can execute the above-mentioned method for making electronic maps, which solves the problem that the production of electronic maps is relatively difficult in the prior art. This application directly obtains road images and road positioning data through road detection tools, then directly identifies the road sign information from the road image, and calculates the actual position of the road sign based on the road positioning data. In this way, the road sign information can be directly mapped to the initial road network generated based on the road positioning data, thereby generating an electronic map without the need for map surveying or data format integration, thus improving the convenience of making electronic maps.

[0110] In the embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0111] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0112] In addition, each functional unit in the embodiments provided by the present application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0113] If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0114] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings, in addition, the terms "first", "second", "third" and the like are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0115] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, used to illustrate the technical solutions of the present application, and not to limit them, the protection scope of the present application is not limited thereto, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: any person skilled in the art within the technical scope disclosed by the present application, they can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. All should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be limited to the protection scope of the claims.

Claims

1. A method of creating an electronic map, characterized by, The method comprises the following steps: obtaining road data detected by a road detection tool in a target area, wherein the road data comprises a plurality of road images and a plurality of road positioning data; for each road image, identifying road indication information of a road sign in the road image, and determining an actual position of the road sign according to road positioning data matched with the road image; wherein the identification of the road indication information of the road sign in the road image comprises the following steps: for each road image, determining a to-be-identified area in the road image according to the position of a horizontal line in the road image; for each road image, determining an effective image area containing the road sign in the to-be-identified area by using an image recognition model; for each road image, determining the road indication information of the road sign in the effective image area by using a character recognition model; generating an initial road network of the target area according to the plurality of road positioning data; adding the road indication information of each road sign to the initial road network according to the actual position of each road sign, to obtain an electronic map of the target area.

2. The method of claim 1, wherein, For each road image, the character recognition model is used to determine the road indication information of the road sign in the effective image area, which comprises the following steps: for each road image, performing deformation correction on the effective image area to obtain a main view of the road sign; for each road image, the character recognition model is used to determine the road indication information of the road sign in the main view of the road sign.

3. The method of claim 2, wherein, For each road image, the deformation correction is performed on the effective image area to obtain the main view of the road sign, which comprises the following steps: for each road image, the image cutting algorithm is used to determine a road sign area and a background area in the effective image area; for each road image, the positions of four vertices of the road sign in the effective image area are determined according to the positions of the boundaries of the road sign area in the effective image area; for each road image, the deformation correction is performed on the road sign area by using the vertex positions of the four vertices of the road sign to obtain the main view of the road sign.

4. The method of claim 1, wherein, For each road image, the actual position of the road sign is determined according to the road positioning data matched with the road image, which comprises the following steps: for each road image, the actual position of the road sign is determined according to the road positioning data matched with the road image and the road positioning data matched with a historical road image; the difference between the shooting time of the historical road image and the shooting time of the road image meets a preset requirement.

5. An electronic map creation device, characterized by comprising: The method comprises the following steps: an acquisition module is configured to obtain road data detected by a road detection tool in a target area, wherein the road data comprises a plurality of road images and a plurality of road positioning data; The determining module is configured to, for each road image, identify road indication information of a road sign in the road image, and determine an actual position of the road sign according to road positioning data matched with the road image; wherein the determining module comprises: a first determining unit configured to, for each road image, determine a to-be-identified region in the road image according to a horizontal line position in the road image; a second determining unit configured to, for each road image, determine an effective image region containing a road sign in the to-be-identified region by using an image recognition model; and a third determining unit configured to, for each road image, determine road indication information of a road sign in the effective image region by using a character recognition model. The generating module is configured to generate an initial road network of the target region according to a plurality of the road positioning data. The adding module is configured to add the road indication information of each road sign to the initial road network according to the actual position of each road sign, to obtain an electronic map of the target region.

6. The apparatus of claim 5, wherein, The third determining unit comprises: A first determining sub-unit configured to, for each road image, perform morphological correction on the effective image region to obtain a front view of the road sign. A second determining sub-unit configured to, for each road image, determine road indication information of a road sign in the front view of the road sign by using a character recognition model.

7. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1-4.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program is run by the processor to implement the steps of the method of any one of claims 1-4.

Images