Dynamic map generation method and system based on multi-dimensional data processing

Abstract

The application provides a dynamic map generation method and system based on multidimensional data processing, and relates to the technical field of data processing.The method comprises the following steps: performing clustering analysis according to target region basic data, generating a data processing channel according to multiple data classes, determining multiple data processing dimensions, respectively performing time point identification on each dimension in the multiple data processing dimensions, generating N identification time point information for the serialization processing of multiple target data, extracting multiple time tracks to determine static information and dynamic information in the target region, and calling the static information and the dynamic information according to the basic map information of the target region to construct a dynamic map of the target region.The application solves the technical problem that the existing technology lacks multidimensional data processing in the process of generating a dynamic map, resulting in low map accuracy, realizes accurate processing of a dynamic map based on multidimensional data, and improves map accuracy.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a method and system for generating dynamic maps based on multidimensional data processing. Background Technology

[0002] With the development of science and technology, especially computer technology, the role and form of maps have changed. These changes have led to a transformation in the development of dynamic map visualization. However, current technologies lack multi-dimensional data processing in the process of generating dynamic maps, resulting in low map accuracy. Summary of the Invention

[0003] This application provides a method and system for generating dynamic maps based on multidimensional data processing, which addresses the technical problem of low map accuracy caused by the lack of multidimensional data processing in the process of generating dynamic maps in the prior art.

[0004] In view of the above problems, this application provides a method and system for generating dynamic maps based on multidimensional data processing.

[0005] Firstly, this application provides a dynamic map generation method based on multidimensional data processing. The method includes: performing cluster analysis on basic data of a target area to generate multiple data classes; generating data processing channels based on the multiple data classes and determining multiple data processing dimensions, wherein the multiple data processing dimensions correspond to the data processing channels; assigning time markers to each of the multiple data processing dimensions to generate N marker time information; performing serialization processing on multiple target data according to the N marker time information to extract multiple time trajectories; determining static and dynamic information in the target area based on the multiple time trajectories; and constructing a dynamic map of the target area by calling the static and dynamic information based on the basic map information of the target area.

[0006] Secondly, this application provides a dynamic map generation system based on multidimensional data processing. The system includes: a clustering analysis module, which performs clustering analysis based on basic data of a target area to generate multiple data classes; a dimension acquisition module, which generates data processing channels based on the multiple data classes and determines multiple data processing dimensions, wherein the multiple data processing dimensions correspond to the data processing channels; a time identification module, which identifies each of the multiple data processing dimensions to generate N identification time information; a serialization processing module, which performs serialization processing on multiple target data according to the N identification time information to extract multiple time trajectories; an information determination module, which determines static and dynamic information in the target area based on the multiple time trajectories; and a map construction module, which constructs a dynamic map of the target area by calling the static and dynamic information based on the basic map information of the target area.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The dynamic map generation method and system based on multidimensional data processing provided in this application relate to the field of data processing technology. It solves the technical problem in the prior art that the lack of multidimensional data processing in the process of generating dynamic maps leads to low map accuracy. It realizes accurate processing of dynamic maps based on multidimensional data and improves map accuracy. Attached Figure Description

[0009] Figure 1 This application provides a schematic diagram of the dynamic map generation method based on multidimensional data processing;

[0010] Figure 2 This application provides a schematic diagram of the structure of a dynamic map generation system based on multidimensional data processing.

[0011] Figure labeling: Clustering analysis module 1, Dimension acquisition module 2, Time marker module 3, Serialization processing module 4, Information determination module 5, Map construction module 6. Detailed Implementation

[0012] This application provides a dynamic map generation method and system based on multidimensional data processing to solve the technical problem in the prior art that the lack of multidimensional data processing in the process of generating dynamic maps leads to low map accuracy.

[0013] Example 1

[0014] like Figure 1As shown in the embodiment of this application, a dynamic map generation method based on multidimensional data processing is provided, the method comprising:

[0015] Step A100: Perform cluster analysis based on the basic data of the target area to generate multiple data classes;

[0016] Furthermore, step A100 of this application also includes:

[0017] Step A110: Traverse the pixels of the region image in the basic data of the target region and obtain the pixel traversal results;

[0018] Step A120: Determine the target pixel density and target pixel chromaticity based on the pixel traversal results;

[0019] Step A130: The target pixel density, the target pixel chromaticity, and the target pixel connectivity are fused to determine multiple fused data;

[0020] Step A140: Perform cluster analysis on the multiple fused data to generate the multiple data types.

[0021] In this application, the dynamic map generation method based on multidimensional data processing provided in the embodiments of this application is applied to a dynamic map generation system based on multidimensional data processing. To generate dynamic maps more accurately based on multidimensional data processing, cluster analysis is first performed on the basic data of the target area. This involves sequentially traversing the pixels of the regional image in the basic data of the target area, identifying the pixel density and pixel chromaticity of each pixel, obtaining the pixel traversal results, and determining the target pixel density and target pixel chromaticity based on the pixel traversal results. This scenario can be set under traffic and road conditions, including traffic and road information, vehicle information, and pedestrian flow information contained in the regional image. The target pixel density refers to the number of pixels per inch of screen. The higher the pixel count and pixel density, the richer the details of the displayed image. Target pixel chromaticity is used to express the color of a pixel. Further, the target pixel density, the target pixel chromaticity, and the target pixel connectivity relationship are fused. Pixel connectivity relationship means that for every value within the target pixel density, there is only one corresponding value for the target pixel chromaticity, while for every value within the target pixel chromaticity, the target pixel density can have multiple corresponding values. The fused data is recorded as multiple fused data sets. Cluster analysis is performed on these multiple fused data sets, grouping the pixel sets into data types composed of similar fused data sets for analysis. Multiple data types are generated, serving as an important reference for the later realization of dynamic map generation based on multidimensional data processing.

[0022] Step A200: Generate data processing channels based on the multiple data types and determine multiple data processing dimensions, wherein the multiple data processing dimensions correspond to the data processing channels;

[0023] Furthermore, step A200 of this application also includes:

[0024] Step A210: Iterate through the data processing methods of the multiple data classes in big data and match them to determine multiple data processing methods;

[0025] Step A220: Train and verify the multiple data processing methods based on the BP neural network to generate a data processing channel;

[0026] Step A230: Generate decomposition rules based on the category thresholds of the multiple data classes;

[0027] Step A240: Decompose the data processing channel based on the decomposition rules to generate the multiple data processing dimensions.

[0028] In this application, to more accurately generate dynamic maps using multidimensional data, it is necessary to construct data processing channels based on the multiple data classes generated above, thereby determining multiple data processing dimensions. These multiple data processing dimensions correspond to the data processing channels. First, the multiple data classes are sequentially used as index data to traverse and access the data processing methods included in the big data. The data processing methods included in the big data can include data collection, data transformation, data grouping, and data organization. Data processing methods with a data matching degree higher than 80% are recorded, thus determining multiple data processing methods. Further, a BP neural network is used to train and validate these multiple data processing methods, generating data processing channels. The training and validation process is as follows:

[0029] The data processing channel is a neural network model in machine learning that can continuously perform self-iterative optimization. The data processing channel is trained using a training dataset and a supervised dataset. Each set of training data in the training dataset includes multiple data processing methods, and the supervised dataset is supervised data that corresponds one-to-one with the training dataset.

[0030] Furthermore, the data processing channel construction process is as follows: each set of training data in the training dataset is input into the data processing channel. The output of the data processing channel is adjusted using the corresponding supervised data. When the output of the data processing channel matches the supervised data, the training of the current set ends. Training is completed on all training data in the training dataset, thus completing the data processing channel training. To ensure the accuracy of the data processing channel, it can be tested using a test dataset. For example, the test accuracy can be set to 80%. When the test accuracy of the test dataset meets 80%, the data processing channel construction is complete.

[0031] The decomposition rule generation based on the category threshold of multiple data classes refers to formulating data class decomposition rules based on the threshold of data state transition between each data class within multiple data classes. Finally, the data processing channel is decomposed according to the decomposition rules, thereby generating multiple data processing dimensions, which ensures the generation of dynamic maps based on multidimensional data processing.

[0032] Step A300: Imprint time information on each of the multiple data processing dimensions to generate N imprint time information;

[0033] In this application, to avoid delays in the dynamic map generated from multidimensional data, it is necessary to sequentially mark the time information of each dimension within the multiple data processing dimensions as determined above. This means marking the time information of the data within the target area processed in each data processing dimension one by one, thereby integrating the data of each data processing dimension after the time marking is completed. That is, the time-marked data in different dimensions is collected, organized, and cleaned, and after data transformation, it is loaded into a new data source. The new data source is then recorded as N marked time information, where N is an integer greater than 2, laying a solid foundation for the subsequent generation of dynamic maps based on multidimensional data processing.

[0034] Step A400: Perform serialization processing on multiple target data according to the N identifier time information to extract multiple time trajectories;

[0035] Furthermore, step A400 of this application also includes:

[0036] Step A410: Perform weight training on the basic data of the target region and determine the weight training results;

[0037] Step A420: Based on the weight training results, extract data from the basic data of the target region to determine the multiple target data;

[0038] Step A430: Connect and sort the N identifier time information according to the time order to generate an identifier time sequence;

[0039] Step A440: Serialize the multiple target data according to the identified time sequence, and connect the serialization results to generate the multiple time trajectories.

[0040] In this application, to ensure the accuracy of the dynamically generated map, it is necessary to serialize multiple target data according to N identifier time information. This means firstly, weight training is performed on the basic data of the target area. The basic target data may include vehicle information, road information, etc. Weight training requires a large amount of data aggregation and precise determination of weights before targeted calculation. For example, the weight ratio of vehicle information and road information can be 4:6 (first influence coefficient: second influence coefficient). Then, the influence parameters after the weight training process are first influence parameter * 0.4 and second influence parameter * 0.6, respectively. Matching is obtained based on this weight training result. The weight training results are then used to extract data from the target area's basic data. This involves determining and extracting target data according to their weights from largest to smallest. The extracted target data with weights greater than 80% are then summarized and recorded as multiple target data sets. Furthermore, the N identifier time information are sequentially connected and sorted according to the time process to generate an identifier time sequence. Finally, the multiple target data sets are serialized according to the identifier time sequence, and the serialization results are connected to generate multiple time trajectories. This process plays a limited role in generating dynamic maps based on multidimensional data processing.

[0041] Step A500: Determine the static and dynamic information in the target area based on the multiple time trajectories;

[0042] Furthermore, step A500 of this application also includes:

[0043] Step A510: Preset state threshold;

[0044] Step A520: Determine whether the state information of the target area in the multiple time trajectories is less than or equal to the state threshold value;

[0045] Step A530: If it is less than or equal to, then the data in the first region is identified as the static information;

[0046] Step A540: If it is greater than, then the second region data is identified as the dynamic information, wherein the first region data and the second region data are determined by randomly selecting multiple region data in the target region, and the first region data and the second region data are different data.

[0047] In this application, the extracted time trajectories are used as basic data to determine the static and dynamic information in the target area. This means that firstly, a state threshold is preset based on the conversion value between historical static and historical dynamic information in the target area. Further, it is determined whether the state information of the target area in the multiple time trajectories is less than or equal to the state threshold. If the state information of the target area in the multiple time trajectories is less than or equal to the state threshold, the corresponding time trajectory is extracted, and the first area data under that time trajectory is marked as static information. If the state information of the target area in the multiple time trajectories is greater than the state threshold, the corresponding time trajectory is extracted, and the second area data is marked as dynamic information. The first area data and the second area data are randomly selected from multiple area data in the target area, and the first area data and the second area data are different data. This determines the static and dynamic information in the target area, so as to serve as reference data for the subsequent generation of dynamic maps based on multidimensional data processing.

[0048] Step A600: Construct a dynamic map of the target area by calling the static information and the dynamic information based on the basic map information of the target area.

[0049] Furthermore, step A600 of this application also includes:

[0050] Step A610: Construct a three-dimensional coordinate system for the target area and generate the basic map information;

[0051] Step A620: Synchronize the static information and the dynamic information to the base map information for dynamic change analysis to obtain an initial dynamic map of the region;

[0052] Step A630: Preset a dynamic monitoring period based on the multiple time trajectories;

[0053] Step A640: Based on the preset dynamic monitoring cycle, perform anomaly monitoring on the static information and dynamic information in the initial dynamic map of the area, and generate an anomaly cycle monitoring file;

[0054] Step A650: Update the initial dynamic map of the area according to the abnormal periodic monitoring file to generate the dynamic map.

[0055] In this application, a dynamic map of the target area is constructed by calling the aforementioned static and dynamic information based on the base map information of the target area. First, based on the space of the target area, a three-dimensional coordinate system of the target area can be constructed, with the y-axis running from south to north, the x-axis running from west to east, and the z-axis running upwards from the horizon at the original point where the x-axis and y-axis intersect. Based on this, information searched within the area image is synchronized to the three-dimensional coordinate system of the target area, thereby generating base map information. Furthermore, static and dynamic information are synchronized to the base map information for dynamic change analysis. Dynamic change analysis is performed when static information undergoes state transitions, and continuous state dynamic change analysis is performed on dynamic information, thus completing the acquisition of the initial dynamic map of the area. Simultaneously, a preset dynamic monitoring period is established for the static and dynamic information of the target area based on the multiple time trajectories extracted above. The preset dynamic monitoring cycle is obtained by pre-setting the dynamic change time nodes in the dynamic change analysis. Furthermore, based on the preset dynamic monitoring cycle, anomaly monitoring is performed on the static and dynamic information in the initial dynamic map of the region. This anomaly monitoring can include the observation that when the fixed static data in the static information and the fixed dynamic information in the dynamic information change, it is considered an anomaly. For example, the road information in the target area is fixed static data. When the road information changes, it is considered abnormal data. Finally, all abnormal data are integrated to generate an anomaly cycle monitoring file. Finally, the initial dynamic map of the region is updated based on the anomaly cycle monitoring file. This means that the abnormal data in the initial dynamic map of the region is removed and the dynamic map information is updated at the same time. Then, the updated map is recorded as a dynamic map and output, which improves the accuracy of generating dynamic maps based on multidimensional data processing in the later stage.

[0056] In summary, the dynamic map generation method based on multidimensional data processing provided in this application has at least the following technical effects: it achieves accurate processing of dynamic maps based on multidimensional data and improves map accuracy.

[0057] Example 2

[0058] Based on the same inventive concept as the dynamic map generation method based on multidimensional data processing in the foregoing embodiments, such as Figure 2 As shown, this application provides a dynamic map generation system based on multidimensional data processing, the system comprising:

[0059] Clustering analysis module 1, which is used to perform clustering analysis based on the basic data of the target area to generate multiple data classes;

[0060] Dimension acquisition module 2 is used to generate data processing channels based on the multiple data classes and determine multiple data processing dimensions, wherein the multiple data processing dimensions and the data processing channels have a corresponding relationship;

[0061] Time identification module 3 is used to identify the time for each of the multiple data processing dimensions and generate N time identification information;

[0062] Serialization processing module 4 is used to perform serialization processing on multiple target data according to the N identifier time information and extract multiple time trajectories;

[0063] Information determination module 5, the information determination module 5 is used to determine static information and dynamic information in the target area based on the multiple time trajectories;

[0064] Map building module 6 is used to construct a dynamic map of the target area by calling the static information and the dynamic information based on the basic map information of the target area.

[0065] Furthermore, the system also includes:

[0066] The first traversal module is used to traverse the pixels of the region image in the basic data of the target region and obtain the pixel traversal results.

[0067] The second traversal module is used to determine the target pixel density and target pixel chromaticity based on the pixel traversal results.

[0068] A fusion module is used to fuse the target pixel density, the target pixel chromaticity, and the target pixel connection relationship to determine multiple fused data.

[0069] A clustering module is used to perform clustering analysis on the multiple fused data to generate the multiple data types.

[0070] Furthermore, the system also includes:

[0071] The matching module is used to traverse the data processing methods of the multiple data classes in big data and match them to determine the multiple data processing methods;

[0072] The training and verification module is used to train and verify the multiple data processing methods based on a BP neural network, and generate a data processing channel.

[0073] The first decomposition module is used to generate decomposition rules based on the category thresholds of the plurality of data classes;

[0074] The second decomposition module is used to decompose the data processing channel based on the decomposition rules to generate the multiple data processing dimensions.

[0075] Furthermore, the system also includes:

[0076] A weight training module is used to train weights on basic data of the target region and determine the weight training results.

[0077] The data extraction module is used to extract data from the basic data of the target region based on the weight training results, and to determine the plurality of target data.

[0078] A connection sorting module is used to connect and sort the N identifier time information according to the time order to generate an identifier time sequence.

[0079] A serialization processing module is used to serialize the multiple target data according to the identified time sequence, and connect the serialization processing results to generate the multiple time trajectories.

[0080] Furthermore, the system also includes:

[0081] A critical value determination module is used to preset state critical values;

[0082] The first judgment module is used to determine whether the state information of the target area in the plurality of time trajectories is less than or equal to the state threshold value.

[0083] The second judgment module is used to identify the first region data as the static information if it is less than or equal to the specified value.

[0084] The third judgment module is used to identify the second region data as the dynamic information if it is greater than the target region data. The first region data and the second region data are determined by randomly selecting multiple region data in the target region, and the first region data and the second region data are different data.

[0085] Furthermore, the system also includes:

[0086] A coordinate system construction module is used to construct a three-dimensional coordinate system for the target area and generate the basic map information;

[0087] A dynamic change analysis module is used to synchronize the static information and the dynamic information to the base map information for dynamic change analysis to obtain an initial dynamic map of the region.

[0088] The period module is used to preset a dynamic monitoring period based on the plurality of time trajectories;

[0089] Anomaly monitoring module, which is used to perform anomaly monitoring on the static information and dynamic information in the initial dynamic map of the area based on the preset dynamic monitoring period, and generate anomaly period monitoring file;

[0090] An update module is used to update the initial dynamic map of the area based on the abnormal periodic monitoring archive, thereby generating the dynamic map.

[0091] Through the foregoing detailed description of the dynamic map generation method based on multidimensional data processing, those skilled in the art can clearly understand the dynamic map generation system based on multidimensional data processing in this embodiment. As for the apparatus disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section description.

[0092] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A dynamic map generation method based on multidimensional data processing, characterized in that, The method includes: Cluster analysis is performed based on basic data of the target region to generate multiple data classes; the method for performing cluster analysis based on basic data of the target region to generate multiple data classes includes: Traverse the pixels of the region image in the basic data of the target region and obtain the pixel traversal results; The target pixel density and target pixel chromaticity are determined based on the pixel traversal results. The target pixel density, the target pixel chromaticity, and the target pixel connectivity are fused to determine multiple fused data. Cluster analysis is performed on the multiple fused data sets to generate the multiple data types; Data processing channels are generated based on the multiple data types, and multiple data processing dimensions are determined. The method for generating data processing channels and determining multiple data processing dimensions based on multiple data types includes: The data processing methods of the multiple data classes in big data are matched to determine multiple data processing methods; The multiple data processing methods are trained and verified based on a BP neural network to generate a data processing channel. Decomposition rules are generated based on the category thresholds of the multiple data classes; The data processing channel is decomposed based on the decomposition rules to generate the multiple data processing dimensions; The multiple data processing dimensions correspond to the data processing channels; Each of the multiple data processing dimensions is marked with a time identifier, generating N time identifier information; Serialization processing of multiple target data is performed based on the N identifier time information to extract multiple time trajectories; Static and dynamic information in the target area are determined based on the multiple time trajectories. A dynamic map of the target area is constructed by calling the static information and the dynamic information based on the base map information of the target area.

2. The method as described in claim 1, characterized in that, The method for serializing multiple target data according to the N identifier time information to extract multiple time trajectories includes: Weight training is performed on the basic data of the target area to determine the weight training results; Based on the weight training results, data is extracted from the basic data of the target region to determine the multiple target data; The N identifier time information are connected and sorted according to the time order to generate an identifier time sequence; The multiple target data are serialized according to the identified time sequence, and the serialization results are connected to generate the multiple time trajectories.

3. The method as described in claim 1, characterized in that, The method for determining static and dynamic information in the target area based on the multiple time trajectories includes: Preset state threshold; Determine whether the state information of the target region in the multiple time trajectories is less than or equal to the state threshold value; If it is less than or equal to, then the data in the first region is identified as the static information; If the value is greater than the target area, the second area data is identified as the dynamic information. The first area data and the second area data are determined by randomly selecting multiple area data from the target area, and the first area data and the second area data are different data.

4. The method as described in claim 1, characterized in that, The method for constructing a dynamic map of the target area by calling the static information and the dynamic information based on the base map information of the target area includes: Construct a three-dimensional coordinate system for the target area and generate the basic map information; The static information and the dynamic information are synchronized to the base map information for dynamic change analysis to obtain an initial dynamic map of the region. A dynamic monitoring cycle is preset based on the multiple time trajectories; Based on the preset dynamic monitoring cycle, anomaly monitoring is performed on the static and dynamic information in the initial dynamic map of the region to generate an anomaly cycle monitoring file. The initial dynamic map of the region is updated based on the abnormal periodic monitoring archive to generate the dynamic map.

5. A dynamic map generation system based on multidimensional data processing, characterized in that, The system includes: The clustering analysis module is used to perform clustering analysis based on the basic data of the target region to generate multiple data classes. It is also used to traverse the pixels of the region image in the basic data of the target region and obtain the pixel traversal results. The target pixel density and target pixel chromaticity are determined based on the pixel traversal results. The target pixel density, the target pixel chromaticity, and the target pixel connectivity are fused to determine multiple fused data. Cluster analysis is performed on the multiple fused data sets to generate the multiple data types; The dimension acquisition module is used to generate data processing channels based on the multiple data classes and determine multiple data processing dimensions, wherein the multiple data processing dimensions and the data processing channels have a corresponding relationship; the module is also used to traverse the data processing methods of the multiple data classes in big data for matching and determine multiple data processing methods; The multiple data processing methods are trained and verified based on a BP neural network to generate a data processing channel. Decomposition rules are generated based on the category thresholds of the multiple data classes; The data processing channel is decomposed based on the decomposition rules to generate the multiple data processing dimensions; A time identification module is used to identify the time for each of the multiple data processing dimensions, generating N time identification information; A serialization processing module is used to perform serialization processing on multiple target data according to the N identifier time information, and extract multiple time trajectories; An information determination module is used to determine static and dynamic information in a target area based on the multiple time trajectories. A map building module is used to construct a dynamic map of the target area by calling the static information and the dynamic information based on the basic map information of the target area.

Images