Method for processing map change information and related device

Abstract

The application relates to the technical field of maps, and provides a map change information processing method and related equipment. The map change information processing method comprises the following steps: acquiring road change detection data, wherein the road change detection data is real-time detection data of a road that has a change relative to a preset map; performing credibility evaluation on the road change detection data; if the credibility evaluation passes, performing change type analysis and data integration on the road change detection data to obtain to-be-processed data; and generating a map change report based on the to-be-processed data. The method improves the reliability and efficiency of map updating.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of map, in particular to a map change information processing method and related equipment. BACKGROUND

[0002] With the development of positioning and navigation technology, the accuracy of navigation depends on the accuracy of the map, so real-time updating of the map is an important content. The target is generally collected by special vehicles to collect road data, generate corresponding maps according to the road data collected by special vehicles, manually label the generated maps to obtain map change information, and update the map.

[0003] The above method can update the map according to the map change information to a certain extent, but the reliability and efficiency of the map updating is low. SUMMARY

[0004] The present application provides a map change information processing method and related equipment, which improves the reliability and efficiency of map updating.

[0005] The first aspect of the present application provides a map change information processing method, comprising:

[0006] Obtain road change detection data, wherein the road change detection data is real-time detection data of the road that has changed relative to the preset map;

[0007] Conduct credibility evaluation on the road change detection data;

[0008] If the credibility evaluation is passed, conduct change type analysis and data integration on the road change detection data to obtain the to-be-processed data;

[0009] Generate a map change report based on the to-be-processed data.

[0010] The second aspect of the present application provides a map change information processing method, comprising:

[0011] Obtain a map change report and visualize the map change report, wherein the map change report is a report generated based on the to-be-processed data corresponding to the road change detection data;

[0012] In response to the acceptance change instruction of the visualized map change report, generate acceptance feedback information;

[0013] In response to the editing instruction of the visualized map change report, generate adjustment information based on the target label information adjusted in the map change report;

[0014] In response to the rejection change instruction of the visualized map change report, generate rejection feedback information.

[0015] The third aspect of the present application provides a processing method of map change information, comprising:

[0016] acquiring road data in real time through a preset data acquisition device to obtain road real-time detection data;

[0017] comparing and analyzing the road real-time detection data with a preset map to obtain a comparison and analysis result;

[0018] if the comparison and analysis result indicates that there is a change, determining the road real-time detection data as road change detection data.

[0019] The fourth aspect of the present application provides a server, comprising:

[0020] an acquisition module configured to acquire road change detection data, wherein the road change detection data is real-time detection data of a road that has a change relative to a preset map;

[0021] an evaluation module configured to perform credibility evaluation on the road change detection data;

[0022] an integration module configured to, if the credibility evaluation passes, perform change type analysis and data integration on the road change detection data to obtain to-be-processed data;

[0023] a first generation module configured to generate a map change report based on the to-be-processed data.

[0024] The fifth aspect of the present application provides a terminal, comprising:

[0025] a visualization module configured to acquire a map change report and visualize the map change report, wherein the map change report is a report generated based on to-be-processed data corresponding to road change detection data;

[0026] a second generation module configured to generate acceptance feedback information in response to an acceptance change instruction of the visualized map change report;

[0027] a third generation module configured to generate adjustment information adjusted based on target annotation information in the map change report in response to an editing instruction of the visualized map change report;

[0028] a fourth generation module configured to generate rejection feedback information in response to a rejection change instruction of the visualized map change report.

[0029] The sixth aspect of the present application provides a terminal, comprising:

[0030] an acquisition module configured to acquire road data in real time through a preset data acquisition device to obtain road real-time detection data;

[0031] The analysis module is configured to compare the road real-time detection data with a preset map to obtain a comparison result.

[0032] The determination module is configured to determine the road real-time detection data as road change detection data if the comparison result indicates that there is a change.

[0033] The seventh aspect of the present application provides a map change information processing device, comprising a memory and at least one processor, the memory stores a computer program; the at least one processor calls the computer program in the memory, so that the map change information processing device executes the above-mentioned map change information processing method.

[0034] The eighth aspect of the present application provides a computer readable storage medium, the computer readable storage medium stores a computer program, when it runs on a computer, makes the computer execute the above-mentioned map change information processing method.

[0035] In the technical solution provided by the present application, road change detection data is obtained, wherein the road change detection data is real-time detection data of a road that has changed relative to a preset map; the road change detection data is subjected to credibility evaluation; if the credibility evaluation passes, the road change detection data is subjected to change type analysis and data integration to obtain to-be-processed data; and a map change report is generated based on the to-be-processed data. In the embodiment of the present application, the credibility evaluation of the road change detection data is performed for secondary confirmation, and the change type analysis and data integration are performed to generate the map change report, thereby ensuring the quality of the road change detection data, improving the effectiveness of the map change information, and further improving the reliability of map updating.

[0036] In the technical solution provided by the present application, a map change report is obtained, and the map change report is subjected to visualization, wherein the map change report is a report generated based on to-be-processed data corresponding to road change detection data; in response to an acceptance change instruction of the visualized map change report, acceptance feedback information is generated; in response to an editing instruction of the visualized map change report, adjustment information adjusted based on target annotation information in the map change report is generated; and in response to a rejection change instruction of the visualized map change report, rejection feedback information is generated. In the embodiment of the present application, the map change report is subjected to visualization, auditing and editing, thereby achieving auditing and adjustment of target annotation information generated based on to-be-processed data corresponding to road change detection data, improving the accuracy and reliability of the target annotation information, and further improving the reliability and efficiency of map updating.

[0037] The technical scheme provided by the present application comprises the following steps: collecting road data in real time through a preset data collection device to obtain road real-time detection data; comparing and analyzing the road real-time detection data with a preset map to obtain a comparison and analysis result; and determining the road real-time detection data as road change detection data if the comparison and analysis result indicates that there is a change. In the embodiment of the present application, the collected road real-time detection data is compared and analyzed with the preset map, thereby solving the limitation of the collection condition of the road real-time detection data, improving the efficiency of the collection of the road real-time detection data, and preliminarily ensuring the quality of the road real-time detection data. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 An embodiment schematic diagram of the map change information processing method in the embodiment of the present application;

[0039] Figure 2 Another embodiment schematic diagram of the map change information processing method in the embodiment of the present application;

[0040] Figure 3 An embodiment schematic diagram of the original annotation and the new annotation in the embodiment of the present application;

[0041] Figure 4 An embodiment schematic diagram of the target visualization type corresponding to the change type information in the embodiment of the present application;

[0042] Figure 5 An embodiment schematic diagram of the server in the embodiment of the present application;

[0043] Figure 6 An embodiment schematic diagram of the terminal in the embodiment of the present application;

[0044] Figure 7 An embodiment schematic diagram of the terminal in the embodiment of the present application. DETAILED DESCRIPTION

[0045] The embodiment of the present application provides a map change information processing method and related equipment, and improves the efficiency and reliability of map updating.

[0046] The terms "first", "second", "third", "fourth" and the like in the description and in the claims of the present application, and above-described drawings, if any, are used to distinguish between similar objects and are not necessarily used to describe a particular sequential or chronological order. It is to be understood that the use of the terms so-termed, if any, can be interchanged, where appropriate, to refer to a similar object in order to enable the embodiments described herein to be carried out in other than the order shown or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or apparatus that comprises a list of steps or units not necessarily limited to those explicitly stated, but can include other steps or units not expressly listed or inherent to such process, method, product or apparatus.

[0047] Based on the above, the embodiments of the present application provide a map change information processing method and related equipment. Mainly applied in the field of automatic driving, not limited here.

[0048] It can be understood that the execution subject of the present application can be a terminal or a server, which is not limited here.

[0049] For ease of understanding, the specific process of the embodiments of the present application is described below. Please refer to Figure 1 An embodiment of the map change information processing method in the embodiments of the present application includes:

[0050] 101、The first terminal collects road data in real time through a preset data acquisition device to obtain road real-time detection data;

[0051] For example, the data acquisition device is used to capture the characteristics, signs, markings and other environmental data of the road in real time. The number of data acquisition devices is one or more, the number of types of data acquisition devices is one or more, and the number of each type of data acquisition device is one or more. The first terminal is a terminal with map navigation function loaded with a preset data acquisition device, such as a wearable device or a riding navigation device. The first terminal can also be a terminal on a device with map navigation function loaded with a preset data acquisition device, such as a vehicle. For example, the road real-time detection data includes at least one of position positioning information, radar point cloud data, video data and image data.

[0052] Optionally, the first terminal collects road data in a preset range in real time through a preset data acquisition device to obtain road real-time detection data. The preset range can be understood as a range corresponding to a semicircle or a full circle with the first terminal as the center and a preset distance as the radius.

[0053] 102. The first terminal compares and analyzes the real-time road detection data with the preset map to obtain the comparison and analysis results;

[0054] As an example and not a limitation, the aforementioned preset map is a map based on precise positioning and detailed data, which can provide accurate and detailed road network information, three-dimensional building models with realistic height and shape, and real-time traffic flow data.

[0055] Optionally, the first terminal converts the real-time road detection data based on the data format of a preset map to obtain the converted real-time road detection data; based on a preset target model, the converted real-time road detection data is compared and analyzed with the preset map to obtain the analysis results.

[0056] As an example, and not a limitation, the process involves calling a pre-defined target model, extracting location information from the real-time road detection data to obtain map information corresponding to the location information, extracting features from the converted real-time road detection data and the map information corresponding to the location information based on the pre-defined target model, and obtaining real-time road detection features and map features. The real-time road detection features and map features are then matched (e.g., similarity calculation), and an analysis result is generated based on the matching result. For example, if the similarity of the features is greater than a preset value, an analysis result is generated, indicating that there has been no change.

[0057] Optionally, if the analysis results indicate no change, the real-time road detection data in the first terminal is deleted, or the real-time road detection data is transferred to a preset storage space.

[0058] 103. If the comparison analysis results indicate a change, the first terminal will determine the real-time road detection data as the road change detection data;

[0059] Optionally, if the comparative analysis results indicate a change, the first terminal will delete duplicate data from the real-time road detection data and perform data integrity and security checks on the deleted real-time road detection data. If the checks pass, the deleted real-time road detection data will be identified as road change detection data. This improves the reliability of road change detection data.

[0060] After the first terminal obtains the road change detection data, it sends the road change detection data to...

[0061] The first terminal compares and analyzes the collected real-time road detection data with a preset map, which solves the limitations of the collection conditions for real-time road detection data, improves the efficiency of real-time road detection data collection, and initially ensures the quality of real-time road detection data.

[0062] 104、the server acquires road change detection data, wherein the road change detection data is real-time detection data of a road that has changed relative to a preset map;

[0063] The server receives the road change detection data sent by the first terminal in real time. The server has strong computing performance, strong network performance, an efficient storage system, and a large memory capacity.

[0064] Optionally, after receiving the road change detection data, the server caches the road change detection data to improve the processing capacity and efficiency of the road change detection data. Alternatively, the server stores the road change detection data to a disaster recovery center for data backup.

[0065] 105、the server performs credibility evaluation on the road change detection data;

[0066] By way of example and not limitation, the server performs credibility evaluation on the road change detection data to again detect / confirm whether the road real-time detection data has indeed changed relative to the preset map, so as to perform secondary data screening and confirmation on the road change detection data and improve the quality and reliability of the road change detection data.

[0067] The content of the credibility evaluation performed by the server on the road change detection data includes but is not limited to performance evaluation of a model corresponding to the road change detection data, positioning signal strength evaluation, data integrity (e.g., whether there is missing or abnormality, whether image or video content is blocked), and data consistency (e.g., whether there is contradiction or inconsistency in multiple detection results of the same road segment, whether there is multiple reporting of the first terminal or reporting of multiple first terminals of the same road segment).

[0068] Optionally, the server can perform multiple credibility evaluation algorithms (one credibility evaluation algorithm corresponds to one credibility evaluation content) in parallel to perform credibility evaluation on the road change detection data, to obtain evaluation results of the multiple credibility evaluation algorithms; if the evaluation results of the multiple credibility evaluation algorithms all meet the respective preset conditions, it is determined that the credibility evaluation passes; if the evaluation results of the multiple credibility evaluation algorithms contain at least one evaluation result that does not meet the respective preset conditions, it is determined that the credibility evaluation does not pass.

[0069] 106、if the credibility evaluation passes, the server performs change type analysis and data integration on the road change detection data to obtain to-be-processed data;

[0070] Optionally, if the credibility evaluation passes, the server acquires a data type in the road change detection data, wherein the data type includes but is not limited to an image, a video, and a point cloud; calls an algorithm corresponding to the data type to identify corresponding data in the road change detection data, and obtains an identification result, wherein the identification result is a type of an identified object; determines a change type according to the identification result, which includes but is not limited to, as an example but not limitation, a newly added traffic cone, a traffic barrier, a construction area, a road pothole, a stop line change, a traffic light position change, a traffic light type change, a road marking change, and a pothole; generates corresponding labeling information according to the change type, wherein the labeling information includes but is not limited to original labeling on a preset map and new labeling generated according to the change type; performs key frame extraction on the data in the road change detection data, and obtains key frame information; merges repeated data in the road change detection data or the key frame information, and obtains merged data; and processes the change type, the labeling information, the key frame information, and the merged data as the to-be-processed data.

[0071] Optionally, the number of identification results of the same identified object is more than two, that is, one algorithm corresponds to one identification result, and the determination of the change type according to the identification result includes: if all the identification results of the same identified object are the same, the identification result is taken as the change type; if all the identification results of the same identified object contain different identification results, the weight of the same identification result and the weight of the different identification result are calculated, and the identification result with a larger weight value is taken as the change type.

[0072] 107. The server generates a map change report based on the to-be-processed data.

[0073] Optionally, the server reads a report template; writes the to-be-processed data into the report template, and obtains the map change report.

[0074] Further, the server can perform corresponding conversion and classification on the to-be-processed data according to each item of the report template, obtain classified data, and write the classified data into the report template to obtain the map change report.

[0075] After the server generates the map change report based on the to-be-processed data, the server sends the map change report to a second terminal.

[0076] The server performs secondary confirmation through credibility evaluation of the road change detection data, analyzes the change type, integrates the data, and generates the map change report, thereby ensuring the quality of the road change detection data, improving the effectiveness of the map change information, and further improving the reliability of the map update.

[0077] 108、the second terminal acquires the map change report, and visualizes the map change report, wherein the map change report is a report generated based on the to-be-processed data corresponding to the road change detection data;

[0078] Optionally, the second terminal receives the map change report sent by the server in real time, and visualizes the map change report according to a data type.

[0079] Optionally, the corresponding data in the map change report is extracted according to a preset visualization form, and the extracted data is visualized according to the preset visualization form.

[0080] The second terminal visualizes the map change report, so as to facilitate intuitive understanding and convenient editing of the road change detection data by the annotator.

[0081] When the annotator audits the visualized map change report, if the automatic annotation suggestion in the map change report is accepted, a corresponding button on a visualization interface or a preset window of the second terminal is clicked to trigger an acceptance change instruction of the visualized map change report; if editing is needed, a corresponding program or tool supported by the second terminal is clicked to adjust the automatic annotation, and when the adjustment is completed, a corresponding button on the visualization interface or the preset window of the second terminal is clicked to trigger an editing instruction of the visualized map change report; if the automatic annotation suggestion in the map change report is rejected, a corresponding button on the visualization interface or the preset window of the second terminal is clicked to trigger a rejection change instruction of the visualized map change report.

[0082] 109、the second terminal generates acceptance feedback information in response to the acceptance change instruction of the visualized map change report;

[0083] For example, the acceptance feedback information can be a character or a field or a request for indicating acceptance of the automatic annotation suggestion. The second terminal generates the acceptance feedback information and sends the acceptance feedback information to the server.

[0084] 110、the second terminal generates adjustment information adjusted based on target annotation information in the map change report in response to the editing instruction of the visualized map change report;

[0085] The second terminal acquires editing data generated by a program or a tool for editing in response to the editing instruction of the visualized map change report, and generates adjustment information adjusted based on the target annotation information in the map change report based on the editing data. The second terminal generates the adjustment information adjusted based on the target annotation information in the map change report, and sends the adjustment information adjusted based on the target annotation information in the map change report to the server.

[0086] 111、The second terminal generates a rejection feedback information in response to the visualized map change report.

[0087] As an example but not limitation, the rejection feedback information can be a character or field or request indicating rejection of the automatic labeling suggestion. The second terminal generates the rejection feedback information and sends it to the server.

[0088] The second terminal realizes the audit confirmation and adjustment of the target labeling information generated based on the to-be-processed data corresponding to the road change detection data by visualizing, auditing and editing the map change report, avoids full manual labeling, improves the accuracy and reliability of the target labeling information, and further improves the reliability and efficiency of map updating.

[0089] In the embodiment of the present application, by comparing and analyzing the collected road real-time detection data with the preset map, the limitation of the collection conditions of the road real-time detection data is solved, the efficiency of the road real-time detection data collection is improved, and the quality of the road real-time detection data is preliminarily ensured. The reliability of the road change detection data is evaluated for secondary confirmation, and the change type analysis and data integration are performed to generate a map change report, ensuring the quality of the road change detection data and improving the effectiveness of the map change information. By visualizing, auditing and editing the map change report, the audit confirmation and adjustment of the target labeling information generated based on the to-be-processed data corresponding to the road change detection data are realized, full manual labeling is avoided, the accuracy and reliability of the target labeling information are improved, and the reliability and efficiency of map updating are further improved.

[0090] Please refer to Figure 2 In the embodiment of the present application, the cloud server, the vehicle terminal and the mobile terminal are taken as examples for illustration. Another embodiment of the map change information processing method in the present application includes:

[0091] 201、The vehicle terminal collects road data in real time through the preinstalled data collection device to obtain road real-time detection data;

[0092] As an example but not limitation, the vehicle terminal can be a terminal loaded on a vehicle with an automatic driving system, or a terminal loaded on a vehicle without an automatic driving system, which is not limited in detail. The data collection device includes but is not limited to a camera, a radar and other sensors (such as a global positioning system (GPS), an infrared sensor, an ultrasonic sensor, etc.).

[0093] Optionally, the first terminal collects road data in real time through a preset data collection device; the collected road data is preprocessed, and the preprocessed road data is filtered based on a preset condition to obtain road real-time detection data. As an example but not limitation, the preprocessing of the collected road data includes denoising processing, enhancement processing, stabilization processing, format conversion processing and image frame extraction processing of video data, downsampling processing, filtering processing and format conversion processing of point cloud data, and scaling processing, cropping processing and enhancement processing of image data. The above-mentioned preset condition is a filtering condition corresponding to each data type (for example, video, image and point cloud), for example, the image quality corresponding to the image data satisfies a certain condition (for example, the definition of the image is greater than a preset definition, the information amount is greater than a preset information amount), and repeated, damaged or invalid images need to be removed.

[0094] As an example but not limitation, the above-mentioned road real-time detection data includes but is not limited to positioning information, point cloud data and video data of multi-directional cameras.

[0095] 202、The vehicle-mounted terminal compares and analyzes the road real-time detection data with the preset map to obtain a comparison and analysis result;

[0096] As an example but not limitation, the above-mentioned road real-time detection data includes point cloud data, a preset target model (for example, Iterative Closest Point (ICP) algorithm, Normal Distribution Transform (NDT) algorithm, Normal Distribution Transform-Oriented Moments (NDT-OM) algorithm) is called to match the point cloud data in the road real-time detection data with the point cloud data corresponding to the preset map to determine whether there is a difference between the two, if there is a difference, a comparison and analysis result indicating that there is a change is generated, if there is no difference, a comparison and analysis result indicating that there is no change is generated.

[0097] Or, the road implementation detection data includes video data, the video data in the road implementation detection data is converted into image data, a preset target model (for example, an (Oriented FAST and Rotated BRIEF, ORB) algorithm, a (Binary Robust Invariant Scalable Keypoints, BRISK) algorithm) is called, similarity calculation is performed based on the image data and image data corresponding to the preset map, if the similarity is greater than or equal to a preset threshold, a comparative analysis result indicating that there is no change is generated, and if the similarity is less than the preset threshold, a comparative analysis result indicating that there is a change is generated. Wherein, the target model is a model capable of quickly comparing and analyzing whether the road real-time detection data has a change relative to the preset map.

[0098] Optionally, if the analysis result indicates that there is no change, the road real-time detection data in the vehicle terminal is deleted, or the road real-time detection data is transferred to a preset storage space.

[0099] 203、If the comparative analysis result indicates that there is a change, the vehicle terminal determines the road real-time detection data as road change detection data;

[0100] Optionally, after the vehicle terminal determines the road real-time detection data as road change detection data, the road change detection data is compressed, and the compressed road change detection data is sent to the cloud server, or the vehicle terminal transmits the road change detection data to the cloud server in segments, for example, the vehicle terminal divides the road change detection data into multiple data blocks according to a preset size, and sends the multiple data blocks to the cloud server one by one, or the vehicle terminal converts the road change detection data into a data stream form and sends it to the cloud server in segments. Reducing the transmission time helps to speed up the processing speed of the data and improve the security of the road change detection data.

[0101] The vehicle terminal compares and analyzes the collected road real-time detection data with the preset map, solves the limitation of the collection conditions of the road real-time detection data, improves the efficiency of the road real-time detection data collection, and preliminarily guarantees the quality of the road real-time detection data.

[0102] 204、The cloud server acquires the road change detection data, wherein the road change detection data is real-time detection data of a road that has a change relative to a preset map;

[0103] The cloud server receives the road change detection data sent by the vehicle terminal in real time. Wherein, the cloud server has strong computing performance, strong network performance, efficient storage system and large memory capacity.

[0104] Optionally, after receiving the road change detection data, the cloud server buffers the road change detection data to improve the processing capacity and efficiency of the road change detection data. Alternatively, the cloud server stores the road change detection data to a disaster recovery center for data backup.

[0105] 205. The cloud server scores the target model corresponding to the road change detection data to obtain a model score value, wherein the target model is a model for analyzing whether there is a change in the real-time detection data of the road relative to the preset map.

[0106] Optionally, the cloud server scores the target model corresponding to the road change detection data based on a preset scoring indicator to obtain a model score value, wherein the scoring indicator includes but is not limited to accuracy, precision, recall rate, and F1 score.

[0107] The server scores the target model corresponding to the road change detection data to evaluate the accuracy and reliability of the target model, and further evaluate the accuracy and reliability of the road change detection data processed by the target model.

[0108] 206. The cloud server obtains the positioning signal strength, the detection distance, and the image occlusion area based on the road change detection data, wherein the detection distance is used to indicate the distance between the data collection device and the detected road.

[0109] By way of example, and not limitation, if the road change detection data includes the positioning signal strength and the detection distance fed back by the positioning system (e.g., global positioning system GPS, cellular network, etc.) corresponding to the first terminal, the cloud server can extract the positioning signal strength and the detection distance from the road change detection data. Alternatively, if the road change detection data includes the positioning signal strength, the first terminal location information, and the road information fed back by the positioning system (e.g., global positioning system GPS, cellular network, etc.) corresponding to the first terminal, the cloud server can extract the positioning signal strength from the road change detection data and calculate the detection distance based on the first terminal location information and the road information.

[0110] The image occlusion area is used to indicate the area of the region in which the image information is missing or distorted due to the occlusion of the camera device. Optionally, the road change detection data includes video data, the cloud server converts the video data in the road change detection data into frame image data, and calls a preset recognition algorithm (e.g., edge detection algorithm, segmentation algorithm, or target detection algorithm) to identify the occlusion area of the frame image data and calculate the area of the occlusion area to obtain the image occlusion area.

[0111] 207. The cloud server counts the number of target reporting items based on the road change detection data to obtain a target reporting number. ​​​​​​​

[0112] As an example but not limitation, the target reporting item is the reporting times or the number of reported terminals.

[0113] Optionally, the cloud server can retrieve the preset database or data storage space based on a preset historical period (a preset period before the current time) to obtain reporting historical data corresponding to the road of the road change detection data; perform key information extraction on the reporting historical data and the road change detection data according to the target reporting item to obtain key information corresponding to the target reporting item; and perform quantity statistics on the key information corresponding to the target reporting item to obtain a target reporting quantity.

[0114] 208、The cloud server evaluates at least one of the model score value, the positioning signal strength, the detection distance, the image occlusion area, and the target reporting quantity as a condition for credibility evaluation;

[0115] Optionally, the cloud server performs weighted summation on at least one of the model score value, the positioning signal strength, the detection distance, the image occlusion area, and the target reporting quantity as a condition for credibility evaluation to obtain a credibility value; if the credibility value is greater than a preset target value, the cloud server determines that the credibility evaluation passes; if the credibility value is less than or equal to the preset target value, the cloud server determines that the credibility evaluation does not pass.

[0116] Optionally, the number of conditions for credibility evaluation is one, and if the conditions for credibility evaluation meet the respective corresponding preset conditions, the cloud server determines that the credibility evaluation passes; if the conditions for credibility evaluation do not meet the respective corresponding preset conditions, the cloud server determines that the credibility evaluation does not pass.

[0117] Optionally, the number of conditions for credibility evaluation is more than one, and if each condition for credibility evaluation meets the respective corresponding preset conditions, the cloud server determines that the credibility evaluation passes; if the conditions for credibility evaluation contain a condition for credibility evaluation that does not meet the respective corresponding preset conditions, the cloud server determines that the credibility evaluation does not pass.

[0118] As an example but not limitation, the preset condition corresponding to the model score value is that the model score value is greater than a preset score value; the preset condition corresponding to the positioning signal strength is that the positioning signal strength is greater than a preset strength; the preset condition corresponding to the detection distance is that the detection distance is greater than a preset distance; the preset condition corresponding to the image occlusion area is that the image occlusion area is less than a preset area; and the preset condition corresponding to the target reporting quantity is that the target reporting quantity is greater than a preset reporting quantity.

[0119] By combining the model score value, positioning signal strength, detection distance, image occlusion area and target reporting quantity, secondary screening is performed through multi-angle credibility evaluation, further improving the quality and reliability of the road change detection data.

[0120] 209、If the credibility evaluation passes, the cloud server performs change type analysis and data integration on the road change detection data to obtain to-be-processed data;

[0121] In an implementation manner, the step 209 includes: detecting the road change detection data to obtain change type information; generating corresponding target labeling information based on the change type information and a preset map; performing key frame extraction on point cloud data and video data in the road change detection data based on the change type information to obtain initial detection data, wherein the road change detection data includes the point cloud data and the video data; performing deduplication on the initial detection data to obtain target detection data; and taking the change type information, the target labeling information and the target detection data as the to-be-processed data.

[0122] By way of example, and not limitation, the road change detection data includes video data and point cloud data, and the detection of the road change detection data to obtain the change type information includes: performing image conversion and target frame detection on the video data of the road change detection data to obtain image recognition results of the impassable object; performing object recognition and classification on the point cloud data of the road change detection data to obtain point cloud recognition results of the impassable object; and if the image recognition results of the impassable object and the point cloud recognition results are the same, merging the image recognition results of the impassable object and the point cloud recognition results to obtain the change type information.

[0123] By way of example, and not limitation, the target standard information includes, but is not limited to, original labeling corresponding to the preset map and new labeling generated according to the change type, and can be as shown in Figure 3

[0124] Since the time of the road change detection data is continuous, the point cloud data and the video data of the key change section need to be cropped to facilitate data processing of the second terminal.

[0125] Optionally, the deduplication of the initial detection data to obtain the target detection data includes: classifying the initial detection data according to the same change region to obtain detection data corresponding to each change region; and merging repeated region data in the detection data corresponding to each change region according to coordinates or a map element unique identifier (ID) to obtain the target detection data.

[0126] ​By changing type analysis, generating target annotation information, key frame extraction and deduplication processing, the quality, accuracy, readability and understandability of the to-be-processed data are improved, the subsequent data processing efficiency is improved, and more efficient data storage and management are realized.

[0127] 210. The cloud server generates a map change report based on the to-be-processed data;

[0128] The execution process of this step 210 is similar to the execution process of the above-mentioned step 107, and will not be described here.

[0129] Among them, as an example but not limited to, the above-mentioned map change report can be inserted with video data, image data, point cloud data, and bird's eye view or height map converted from point cloud data.

[0130] The cloud server performs secondary confirmation through the credibility evaluation of the road change detection data, and performs change type analysis and data integration to generate a map change report, ensuring the quality of the road change detection data, improving the effectiveness of the map change information, and further improving the reliability of the map update.

[0131] 211. The mobile terminal obtains the map change report and visualizes the map change report, wherein the map change report is a report generated based on the to-be-processed data corresponding to the road change detection data;

[0132] In an implementation manner, the above-mentioned step 211 includes: determining a corresponding target visualization type based on the change type information of the to-be-processed data corresponding to the road change detection data in the map change report; and visualizing the to-be-processed data based on the target visualization type.

[0133] Among them, as an example but not limited to, the above-mentioned determining a corresponding target visualization type based on the change type information of the to-be-processed data corresponding to the road change detection data in the map change report can be understood as: if the change type information in the to-be-processed data is a traffic light type change or traffic light damage, the target visualization type can be as shown in Figure 4 If the change type information in the to-be-processed data is road marking change or pothole, the corresponding target visualization type is a base map comparison type; if the change type information in the to-be-processed data is signboard or traffic light position change, the corresponding target visualization type is a point cloud comparison type.

[0134] By visualizing according to the target visualization type, the to-be-processed data corresponding to the road change detection data can be effectively displayed, further facilitating intuitive viewing and understanding of the to-be-processed data, enhancing the interactivity and operability of the to-be-processed data, and further facilitating the improvement of the accuracy and efficiency of the to-be-processed data processing.

[0135] In an implementation manner, the visualization of the to-be-processed data based on the target visualization type comprises: if the target visualization type is a video playing type, corresponding video data in the to-be-processed data is displayed in a playing form; if the target visualization type is a base map comparison type, corresponding new base map and historical base map in the to-be-processed data are obtained, and the new base map and the historical base map are displayed in a same display area, wherein the new base map is an aerial view or a height map converted from corresponding point cloud data in the to-be-processed data; if the target visualization type is a point cloud comparison type, corresponding historical point cloud data is obtained, and the historical point cloud data and corresponding point cloud data in the to-be-processed data are displayed in the same display area.

[0136] For example, the aerial view is suitable for judging road marking change, and the height map is suitable for judging pit range. Optionally, the point cloud data in the to-be-processed data is converted into the aerial view or the height map to obtain the new base map; the historical base map corresponding to the road area of the to-be-processed data is extracted from a database; and the new base map and the historical base map are displayed in the same display area according to a preset display position.

[0137] By visualizing according to the target visualization type, the to-be-processed data corresponding to the road change detection data can be effectively displayed, the intuitive observation and understanding of the to-be-processed data are further facilitated, the interactivity and operability of the to-be-processed data are enhanced, and thus the accuracy and efficiency of processing the to-be-processed data are improved.

[0138] 212. The mobile terminal generates acceptance feedback information in response to the change acceptance instruction of the visualized map change report;

[0139] 213. The mobile terminal generates adjusted information based on the target annotation information in the map change report in response to the editing instruction of the visualized map change report;

[0140] 214. The mobile terminal generates rejection feedback information in response to the change rejection instruction of the visualized map change report;

[0141] The execution processes of steps 212-214 are similar to those of steps 109-111, and thus are not described herein.

[0142] 215. The cloud server generates map update information based on the adjusted information based on the target annotation information in the map change report;

[0143] As an example but not limitation, the map update information can be a map patch, which can be understood as a program for repairing known bugs or errors (i.e., parts with changes) in the map. The cloud server sends the map update information to the vehicle terminal after adjusting the map update information based on the adjustment information.

[0144] 216、The cloud server generates the map update information based on the map change report when receiving the acceptance feedback information of the map change report;

[0145] As an example but not limitation, the map update information can be a map patch, which can be understood as a program for repairing known bugs or errors (i.e., parts with changes) in the map. The cloud server sends the map update information to the vehicle terminal after adjusting the map update information based on the adjustment information.

[0146] The cloud server improves the accuracy, completeness, maintainability and scalability of the map update information by generating the map update information based on the adjustment information or the map change report, which helps to improve the efficiency and reliability of the map update.

[0147] 217、The vehicle terminal fuses the map update information with the preset map to obtain an updated map when receiving the map update information, wherein the map update information is update information generated according to the adjustment information adjusted based on the target annotation information in the map change report;

[0148] As an example but not limitation, the map update information can be fused with the preset map to obtain an updated map by using a pre-installed software (such as a geographic information system (GIS)) or a programming language. For example, the map update information can be aligned with the preset map by using GIS software or a programming language. For point data in the map update information, the point data in the map update information can be aligned on the preset map by spatial interpolation or weighted average. For line and surface data in the map update information, the line and surface data in the map update information can be fused into the preset map by superimposition analysis or spatial correlation.

[0149] 218、The vehicle terminal optimizes a preset target model when receiving the rejection feedback information based on the map change report, wherein the target model is a model for analyzing whether there is a change in the real-time detection data of the road relative to the preset map.

[0150] Among them, as an example but not limited to, the above-mentioned optimization of the preset target model includes but is not limited to optimization of parameters (such as learning rate, number of iterations) and optimization of network structure (such as adjustment, increase or decrease of the network framework of the target model).

[0151] Optionally, the to-be-optimized demand is acquired, wherein the to-be-optimized demand can be understood as the content required to be optimized by the target model; a target optimization model (for example, a gradient descent algorithm, an adaptive gradient algorithm) corresponding to the to-be-optimized demand is matched; and the preset target model is optimized by the target optimization model.

[0152] The vehicle terminal can update the map data in real time and automatically by fusing the map update information with the preset map or optimizing the target model, and improve the accuracy of the map data, thereby effectively improving the efficiency and reliability of the map update.

[0153] Among them, since real-time map updating can make the autonomous vehicle adapt to new road conditions and changes faster, improving the reliability and efficiency of map updating is conducive to improving the safety of autonomous driving, and can more accurately identify and avoid potential dangers; since updating the map can help the autonomous vehicle better plan the path, adapt to traffic conditions, accurately locate traffic lights and signs, and perform more accurate positioning, improving the reliability and efficiency of map updating is conducive to improving the performance and feasibility of the autonomous driving system; since real-time map updating can provide more accurate navigation information, making autonomous driving easier, improving the reliability and efficiency of map updating is conducive to improving the comfort and convenience of autonomous driving.

[0154] In the embodiment of the application, by comparing and analyzing the collected real-time road detection data with the preset map, the limitation of the collection conditions of the real-time road detection data is solved, the efficiency of the real-time road detection data collection is improved, and the quality of the real-time road detection data is preliminarily ensured. The quality of the road change detection data is ensured by performing secondary confirmation through credibility evaluation of the road change detection data, performing change type analysis and data integration, and generating a map change report. The effectiveness of the map change information is improved. The target labeling information generated based on the road change detection data is audited, confirmed and adjusted through visualization, auditing and editing of the map change report, full manual labeling is avoided, the accuracy and reliability of the target labeling information are improved, the accuracy, integrity, maintainability and scalability of the map update information are improved by generating the map update information based on the adjustment information or the map change report, the map data can be updated in real time and automatically by fusing the map update information with the preset map or optimizing the target model, and the accuracy of the map data is improved, thereby improving the reliability and efficiency of the map update.

[0155] The processing method of the map change information in the embodiments of the present application is described above, and a server in the embodiments of the present application is described below, please refer to Figure 5 An embodiment of the server in the embodiments of the present application includes:

[0156] The acquisition module 501 is configured to acquire road change detection data, wherein the road change detection data is real-time detection data of a road that has a change relative to a preset map;

[0157] The evaluation module 502 is configured to perform a credibility evaluation on the road change detection data;

[0158] The integration module 503 is configured to, if the credibility evaluation passes, perform a change type analysis and data integration on the road change detection data to obtain to-be-processed data;

[0159] The first generation module 504 is configured to generate a map change report based on the to-be-processed data.

[0160] In the embodiments of the present application, the credibility evaluation on the road change detection data is performed for secondary confirmation, and the change type analysis and data integration are performed to generate the map change report, thereby ensuring the quality of the road change detection data, improving the effectiveness of the map change information, and further improving the reliability of the map update.

[0161] Optionally, the evaluation module 502 can be specifically configured to:

[0162] score a target model in the road change detection data to obtain a model score value, wherein the target model is a model used to analyze whether the real-time detection data of a road has a change relative to a preset map;

[0163] acquire a positioning signal strength, a detection distance, and an image occlusion area based on the road change detection data, wherein the detection distance is used to indicate a distance between a data collection device and a detected road;

[0164] perform a quantity statistics of target reporting items based on the road change detection data to obtain a target reporting quantity;

[0165] evaluate at least one of the model score value, the positioning signal strength, the detection distance, the image occlusion area, and the target reporting quantity as a condition for the credibility evaluation.

[0166] Optionally, the integration module 503 can be specifically configured to:

[0167] perform a non-passable object detection on the road change detection data to obtain change type information, and generate corresponding target labeling information based on the change type information and the preset map;

[0168] extract key frames from the point cloud data and the video data in the road change detection data based on the change type information, to obtain initial detection data, wherein the road change detection data comprises the point cloud data and the video data;

[0169] de-duplicate the initial detection data to obtain target detection data;

[0170] the change type information, the target annotation information, and the target detection data are taken as to-be-processed data.

[0171] Optionally, the server further comprises:

[0172] a fifth generation module 505, configured to generate map update information based on adjustment information when receiving the adjustment information adjusted based on the target annotation information in the map change report;

[0173] a sixth generation module 506, configured to generate map update information based on the map change report when receiving acceptance feedback information of the map change report.

[0174] The functions of the modules and units in the server and the steps in the method for processing map change information correspond to each other, and the functions and implementation processes will not be described here.

[0175] The method for processing map change information in the embodiments of the application is described above, and a terminal in the embodiments of the application is described below. Please refer to Figure 6 One embodiment of the terminal in the embodiments of the application comprises:

[0176] a visualization module 601, configured to acquire a map change report and visualize the map change report, wherein the map change report is a report generated based on to-be-processed data corresponding to road change detection data;

[0177] a second generation module 602, configured to generate acceptance feedback information in response to an acceptance change instruction of the visualized map change report;

[0178] a third generation module 603, configured to generate adjustment information adjusted based on target annotation information in the map change report in response to an editing instruction of the visualized map change report;

[0179] a fourth generation module 604, configured to generate rejection feedback information in response to a rejection change instruction of the visualized map change report.

[0180] In the embodiment of the present application, by visualizing, auditing and editing the map change report, the auditing confirmation and adjustment of the target labeling information generated based on the to-be-processed data corresponding to the road change detection data are realized, the accuracy and reliability of the target labeling information are improved, and the reliability and efficiency of the map updating are improved.

[0181] Optionally, the visualization module 601 can be specifically used for:

[0182] determining the corresponding target visualization type based on the change type information of the to-be-processed data corresponding to the road change detection data in the map change report;

[0183] visualizing the to-be-processed data based on the target visualization type.

[0184] Optionally, the visualization module 601 can be specifically used for:

[0185] if the target visualization type is a video playing type, the corresponding video data in the to-be-processed data is displayed in the form of playing;

[0186] if the target visualization type is a base map comparison type, the corresponding new base map and historical base map in the to-be-processed data are obtained, and the new base map and the historical base map are displayed in the same display area, wherein the new base map is an aerial view or a height map converted from the corresponding point cloud data in the to-be-processed data;

[0187] if the target visualization type is a point cloud comparison type, the historical point cloud data corresponding to the to-be-processed data is obtained, and the historical point cloud data and the corresponding point cloud data in the to-be-processed data are displayed in the same display area.

[0188] The functions of the modules and units in the above server correspond to the steps in the above map change information processing method embodiment, and the functions and implementation processes will not be repeated here.

[0189] The above describes the map change information processing method in the embodiment of the present application, and the following describes another terminal in the embodiment of the present application, please refer to Figure 7 An embodiment of another terminal in the embodiment of the present application includes:

[0190] The acquisition module 701 is configured to acquire road data in real time through a preset data acquisition device to obtain road real-time detection data.

[0191] The analysis module 702 is configured to compare and analyze the road real-time detection data with a preset map to obtain a comparison and analysis result.

[0192] The determination module 703 is configured to determine the road real-time detection data as road change detection data if the comparison and analysis result indicates that there is a change.

[0193] In the embodiment of the present application, by comparing and analyzing the collected road real-time detection data with the preset map, the limitation of the collection condition of the road real-time detection data is solved, the efficiency of the road real-time detection data collection is improved, and the quality of the road real-time detection data is preliminarily ensured.

[0194] Optionally, the terminal further comprises:

[0195] The fusion module 704 is configured to fuse the map update information with the preset map to obtain an updated map when the map update information is received, wherein the map update information is update information generated according to the adjusted information based on the target annotation information in the map change report.

[0196] The optimization module 705 is configured to optimize the preset target model when the rejection feedback information based on the map change report is received, wherein the target model is a model used to analyze whether there is a change in the road real-time detection data relative to the preset map.

[0197] The functions of the modules and units in the server correspond to the steps in the method for processing map change information, and the functions and implementation processes will not be repeated here.

[0198] The present application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium or a volatile computer readable storage medium. The computer readable storage medium stores a computer program. When the computer program runs on a computer, the computer executes the steps of the method for processing map change information.

[0199] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0200] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part 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 plurality of computer programs 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 various embodiment methods of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0201] The above, the above embodiments are only to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for processing map change information, characterized in that, The method for processing map change information includes: Acquire road change detection data, wherein the road change detection data is real-time detection data of roads that have changed relative to a preset map; The reliability of the road change detection data is assessed. If the credibility assessment is passed, then the road change detection data is used to detect impassable objects to obtain change type information; based on the change type information and the preset map, corresponding target labeling information is generated. Based on the change type information, keyframes are extracted from the point cloud data and video data in the road change detection data to obtain initial detection data, wherein the road change detection data includes point cloud data and video data; The initial detection data is deduplicated to obtain the target detection data; The change type information, the target annotation information, and the target detection data are used as data to be processed. A map change report is generated based on the data to be processed.

2. The method for processing map change information according to claim 1, characterized in that, The credibility assessment of the road change detection data includes: The target model corresponding to the road change detection data is scored to obtain a model score value, wherein the target model is a model used to analyze whether there are changes in the real-time road detection data relative to the preset map; Based on the road change detection data, the positioning signal strength, detection distance, and image occlusion area are obtained, wherein the detection distance is used to indicate the distance between the data acquisition device and the detected road; Based on the road change detection data, the number of target reported items is counted to obtain the target reported quantity; The reliability assessment is performed using at least one of the following: the model score, the location signal strength, the detection distance, the area of ​​the image occlusion region, and the number of reported targets.

3. The method for processing map change information according to claim 1, characterized in that, The acquisition of road change detection data includes: Real-time road detection data is obtained by collecting road data in real time through pre-installed data acquisition equipment; The real-time road detection data is compared and analyzed with a preset map to obtain the comparison and analysis results; If the comparative analysis results indicate a change, the real-time road detection data will be identified as road change detection data.

4. The method for processing map change information according to claim 1, characterized in that, After generating the map change report based on the data to be processed, the process also includes: When adjustment information based on the target labeling information in the map change report is received, map update information is generated based on the adjustment information; When the acceptance feedback information of the map change report is received, map update information is generated based on the map change report.

5. A method for processing map change information, characterized in that, The method for processing map change information includes: Obtain a map change report and visualize the map change report, wherein the map change report is obtained by the map change information processing method as described in any one of claims 1-4; In response to the acceptance instruction in the visualized map change report, generate acceptance feedback information; In response to the editing instructions in the visualized map change report, adjustment information is generated based on the target labeling information in the map change report; In response to the rejection instruction in the visualized map change report, a rejection feedback message is generated.

6. The method for processing map change information according to claim 5, characterized in that, The visualization of the map change report includes: The corresponding target visualization type is determined based on the change type information of the pending data corresponding to the road change detection data in the map change report. The data to be processed is visualized based on the target visualization type.

7. The method for processing map change information according to claim 6, characterized in that, The visualization of the data to be processed based on the target visualization type includes: If the target visualization type is a video playback type, then the corresponding video data in the data to be processed will be displayed in the form of playback; If the target visualization type is a base map comparison type, then the corresponding new base map and historical base map in the data to be processed are obtained, and the new base map and the historical base map are displayed in the same display area. The new base map is a bird's-eye view or height map converted from the corresponding point cloud data in the data to be processed. If the target visualization type is a point cloud comparison type, then the historical point cloud data corresponding to the data to be processed is obtained, and the historical point cloud data and the corresponding point cloud data in the data to be processed are displayed in the same display area.

8. The method for processing map change information according to claim 5, characterized in that, The method for processing map change information also includes: When map update information is received, the map update information is merged with a preset map to obtain an updated map. The map update information is generated based on the adjustment information after adjusting the target label information in the map change report. When a rejection feedback message based on a map change report is received, the preset target model is optimized. The target model is a model used to analyze whether there are changes in real-time road detection data relative to the preset map.

9. A server, characterized in that, The server includes: The acquisition module is used to acquire road change detection data, wherein the road change detection data is real-time detection data of roads that have changed relative to a preset map; The evaluation module is used to evaluate the credibility of the road change detection data; An integration module is used to detect impassable objects in the road change detection data if the credibility assessment passes, thereby obtaining change type information; generate corresponding target annotation information based on the change type information and a preset map; extract keyframes from the point cloud data and video data in the road change detection data based on the change type information to obtain initial detection data, wherein the road change detection data includes point cloud data and video data; deduplicate the initial detection data to obtain target detection data; and use the change type information, the target annotation information, and the target detection data as data to be processed. The first generation module is used to generate a map change report based on the data to be processed.

10. A terminal, characterized in that, The terminal includes: A visualization module is used to acquire a map change report and visualize the map change report, wherein the map change report is obtained by the map change information processing method as described in any one of claims 1-4; The second generation module is used to generate acceptance feedback information in response to the acceptance instruction of the visualized map change report; The third generation module is used to generate adjustment information based on the target labeling information in the map change report after the map change report is visualized, in response to the editing instructions of the map change report. The fourth generation module is used to generate rejection feedback information in response to the rejection instruction in the visualized map change report.

11. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the method for processing map change information as described in any one of claims 1-8.

Images