Radar data acquisition method, system and electronic equipment combined with scene error analysis

Abstract

The present invention relates to a radar data collection method, system and electronic device combined with scene error analysis, and to the field of radar data collection technology, including: constructing a three-dimensional terrain model, collecting on-site fog, humidity and temperature data, performing fusion accuracy analysis and fusion processing, and obtaining fog distribution, humidity distribution and temperature distribution; analyzing distribution rationality and detection reproducibility and performing distribution compensation to obtain compensated fog distribution, compensated humidity distribution and compensated temperature distribution; collecting radar data, performing error analysis and correction according to the three-dimensional terrain model, and obtaining first corrected data; performing error analysis and correction according to the compensated fog distribution, compensated humidity distribution and compensated temperature distribution, and obtaining second corrected data, and fusing the two correction results to obtain corrected radar data. The present invention solves the technical problem that the prior art lacks relevant means to effectively fit the disturbance caused by multiple factors, resulting in poor accuracy of radar collection data in bad weather.

Description

Technical Field

[0001] The present invention relates to the field of radar data collection, and in particular to a radar data collection method, system and electronic equipment combined with scene error analysis. Background Art

[0002] Radar is an electronic device that uses electromagnetic waves to detect targets. It obtains information about the target, such as distance, speed, direction, etc., by emitting electromagnetic waves and receiving echoes. Radar can detect distant targets day and night, and is not blocked by fog, clouds and rain. It has the characteristics of all-weather and all-day, and has a certain penetration ability. It is widely used in weather forecasting, aviation, aerospace, navigation, military applications, security, traffic management, and scientific research.

[0003] Weather conditions are one of the main factors that affect radar measurement errors. In bad weather such as fog, rain, and frost, radar signals will be absorbed, scattered, or reflected by raindrops and other particles, resulting in reduced radar measurement accuracy. In addition, changes in wind speed and direction can also affect radar measurement errors.

[0004] The existing technology usually uses noise reduction filtering algorithms, Doppler filtering algorithms and other technologies to process radar signals in rainy and snowy weather conditions, and uses signal reconstruction, target recognition and other methods to improve the detection capability of radar in foggy weather.

[0005] The existing technology lacks relevant means to effectively fit the disturbances caused by multiple factors, resulting in a technical problem of poor accuracy of radar data collection in bad weather. Summary of the invention

[0006] The present invention aims to solve the problem of the influence of weather conditions in the prior art, and the lack of relevant means to effectively fit the disturbance caused by multiple factors, resulting in poor accuracy of radar data collection in bad weather. A radar data collection method, system and electronic equipment combined with scene error analysis are provided.

[0007] In a first aspect, the present invention provides a radar data collection method combined with scene error analysis, including: collecting terrain feature data within a measurement scene to construct a three-dimensional terrain model; collecting fog data, humidity data, and temperature data within the measurement scene, and performing fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data based on historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution; analyzing the distribution rationality of the fog distribution, humidity distribution, and temperature distribution based on the three-dimensional terrain model, and analyzing the detection reproducibility of the fog distribution, humidity distribution, and temperature distribution based on the historical monitoring data of the measurement scene to obtain fog accuracy. The method comprises the following steps: collecting radar data collected by radar measurement in the measurement scene, performing radar data error analysis and correction according to the terrain three-dimensional model, and obtaining first corrected radar data; based on the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, performing radar data error analysis and correction according to the compensated fog distribution, compensated humidity distribution and compensated temperature distribution, and obtaining second corrected radar data, which are then fused with the first corrected radar data to obtain corrected radar data.

[0008] In a second aspect, the present invention provides a radar data acquisition system combined with scene error analysis, including: a terrain data acquisition module, which is used to collect terrain feature data in a measurement scene and construct a three-dimensional terrain model; a data fusion module, which is used to collect fog data, humidity data, and temperature data in the measurement scene, and perform fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution; a distribution compensation module, which is used to analyze the distribution rationality of the fog distribution, humidity distribution, and temperature distribution according to the three-dimensional terrain model, and analyze the detection reproducibility of the fog distribution, humidity distribution, and temperature distribution according to the historical monitoring data of the measurement scene to obtain A fog accuracy coefficient, a temperature accuracy coefficient and a humidity accuracy coefficient are obtained, and the fog distribution, humidity distribution and temperature distribution are compensated to obtain compensated fog distribution, compensated humidity distribution and compensated temperature distribution; a first correction module is used to collect radar data collected by radar measurement in the measurement scene, and perform radar data error analysis and correction according to the three-dimensional terrain model to obtain first corrected radar data; a second correction module is used to perform radar data error analysis and correction based on the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, based on integrated learning, according to the compensated fog distribution, compensated humidity distribution and compensated temperature distribution to obtain second corrected radar data, which are combined with the first corrected radar data for fusion to obtain corrected radar data.

[0009] In a third aspect, the present invention provides an electronic device, comprising: a memory for storing a computer software program; and a processor for reading and executing the computer software program, thereby implementing the radar data collection method combined with scene error analysis as described in any one of the first aspects.

[0010] The beneficial effects of the present invention are: by comprehensively considering the influencing factors of four dimensions, namely terrain, fog, humidity and temperature, and performing multi-dimensional analysis on the errors generated in the four dimensions, a secondary correction of the radar collected data is achieved, and through accuracy analysis and fusion processing, the technical effect of improving the accuracy of radar collected data in severe weather is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic flow chart of a radar data acquisition method combined with scene error analysis provided by the present invention;

[0012] Figure 2 A schematic diagram of the structure of a radar data acquisition system combined with scene error analysis provided by the present invention;

[0013] Figure 3 This is a functional block diagram of the electronic device provided by the present invention.

[0014] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0015] Terrain data acquisition module 1000, data fusion module 2000, distribution compensation module 3000, first correction module 4000, second correction module 5000, electronic device 500, memory 510, processor 520, computer program 511. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0017] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0018] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0019] Example 1

[0020] like Figure 1 As shown, an embodiment of the present invention provides a radar data collection method combined with scene error analysis, including:

[0021] S10: Collect terrain feature data in the measurement scene and construct a three-dimensional terrain model;

[0022] Specifically, the terrain feature data in the measurement scene is obtained through various means such as satellite images, aerial photogrammetry, LiDAR (laser radar) scanning, etc. The terrain three-dimensional model is completed by using the terrain feature data in the measurement scene through professional geographic information system (GIS) software, three-dimensional modeling software and other tools. Specifically, the collected terrain feature data is cleaned, corrected, format converted, etc., and a specific algorithm and software are used to build a network structure of the terrain according to the data, and the collected data is accurately assigned to the network nodes. The model is further image mapped, simplified, smoothed, etc., and visualized using tools such as rendering engines. The terrain three-dimensional model can display terrain elevation data (indicating the undulating shape of the ground), geomorphic feature data (such as the location and shape of mountains, valleys, cliffs, etc.), surface cover data (such as vegetation type, building distribution, etc.) water system data (the location and shape of water systems such as rivers and lakes), road and transportation facilities data, etc. By collecting terrain data and building a three-dimensional model for the measurement scene, data support can be provided for the correction of data errors caused by the terrain.

[0023] S20: collecting fog data, humidity data, and temperature data in the measurement scene, and performing fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution;

[0024] Further, the fog data, humidity data, and temperature data in the measurement scene are collected, and according to the historical monitoring data of the measurement scene, the fog data, humidity data, and temperature data are subjected to fusion accuracy analysis and fusion processing, step S20 includes:

[0025] S21: Collect fog data, humidity data, and temperature data of multiple marked points in the measurement scene;

[0026] S22: constructing a fog fusion branch, a humidity fusion branch, and a temperature fusion branch according to the historical monitoring data of the measurement scene, and performing fusion accuracy analysis to obtain a fog fusion accuracy coefficient set, a humidity fusion accuracy coefficient set, and a temperature fusion accuracy coefficient set, wherein the fog fusion branch includes a first fog distribution construction path, a first humidity distribution construction path, and a first temperature distribution construction path, and the fog fusion accuracy coefficient set includes a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient, and a first temperature distribution accuracy coefficient;

[0027] S23: inputting the fog data, humidity data, and temperature data into the fog fusion branch, humidity fusion branch, and temperature fusion branch respectively, and processing to obtain a fog distribution set, a humidity distribution set, and a temperature distribution set;

[0028] S24: using the fog fusion accuracy coefficient set, humidity fusion accuracy coefficient set and temperature fusion accuracy coefficient set, perform fusion weighted processing on the fog distribution set, humidity distribution set and temperature distribution set to obtain fog distribution, humidity distribution and temperature distribution.

[0029] Specifically, the size of the fog in the air is mainly related to the air humidity and temperature. The greater the air humidity, the more conducive it is to the formation of fog. When the air temperature is high, the amount of water vapor that the air can hold also increases, thereby reducing the formation of fog; when the air temperature is low, the amount of water vapor that the air can hold decreases, and the excess water vapor condenses to form fog. The measurement scene is a real scene where the radar collects data. The fog data, humidity data, and temperature data are obtained by pre-setting multiple marking points in the measurement scene and collecting data through multiple sensors at multiple marking points. At the same time, the meteorological data of the measurement scene area can also be retrieved as auxiliary data to supplement the fog data, humidity data, and temperature data.

[0030] In order to describe the fog distribution, humidity distribution and temperature distribution of the measurement scene, it is necessary to construct the fog fusion branch, humidity fusion branch and temperature fusion branch. Since the fog concentration is correlated with humidity and temperature, the fog fusion branch, humidity fusion branch and temperature fusion branch are respectively composed of three paths. Specifically, the fog fusion branch includes the first fog distribution construction path, the first humidity distribution construction path and the first temperature distribution construction path; the humidity fusion branch includes the second fog distribution construction path, the second humidity distribution construction path and the second temperature distribution construction path; the temperature fusion branch includes the third fog distribution construction path, the third humidity distribution construction path and the third temperature distribution construction path.

[0031] Retrieve the historical monitoring data of the measurement scene, and use the historical monitoring data as the branch training data set. The historical monitoring data includes historical fog data, historical humidity data, historical temperature data, and corresponding fog distribution data, humidity distribution data, and temperature distribution data. Input the historical fog data and fog distribution data, humidity distribution data, and temperature distribution data in the historical monitoring data into the fog fusion branch, and all training data sequentially train the first fog distribution construction path, the first humidity distribution construction path, and the first temperature distribution construction path. When the number of training times meets the preset number of training times, the training of the fog fusion branch is completed. The humidity fusion branch is trained with historical humidity data and fog distribution data, humidity distribution data, and temperature distribution data as training data, and the temperature fusion branch is trained with historical temperature data and fog distribution data, humidity distribution data, and temperature distribution data as training data. The training process is consistent with the training process of the fog fusion branch.

[0032] Further fusion accuracy analysis is performed. Since the fog fusion branch, humidity fusion branch, and temperature fusion branch have different processing accuracy for fog data, humidity data, and temperature data, it is necessary to perform accuracy analysis on the fog fusion branch, humidity fusion branch, and temperature fusion branch. The accuracy analysis can be judged based on the difference between the predicted distribution data and the historical distribution data output during the fusion branch training process. For example, if the temperature distribution difference is 20%, the temperature distribution accuracy coefficient is 80%, and the sum of the two is 100%.

[0033] The fog fusion accuracy coefficient set includes a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient and a first temperature distribution accuracy coefficient. The humidity fusion accuracy coefficient set includes a second fog distribution accuracy coefficient, a second humidity distribution accuracy coefficient and a second temperature distribution accuracy coefficient. The temperature fusion accuracy coefficient set includes a third fog distribution accuracy coefficient, a third humidity distribution accuracy coefficient and a third temperature distribution accuracy coefficient. After the fog fusion branch, the humidity fusion branch and the temperature fusion branch are trained and analyzed for fusion accuracy, the fog data, humidity data and temperature data are respectively input into the fog fusion branch, the humidity fusion branch and the temperature fusion branch. After the branches process the data, the fog distribution set, the humidity distribution set and the temperature distribution set are output.

[0034] The fog distribution set, humidity distribution set and temperature distribution set are subjected to the fusion weighted processing, and the first fog distribution accuracy coefficient, the second fog distribution accuracy coefficient and the third fog distribution accuracy coefficient are used as the weight values ​​of the fog distribution set fusion weighting, and the fog distribution is obtained by multiplying the weights of each point data of the fog distribution set. Similarly, the first humidity distribution accuracy coefficient, the second humidity distribution accuracy coefficient and the third humidity distribution accuracy coefficient are used as the weight values ​​of the fog distribution set fusion weighting, and the humidity distribution is obtained by multiplying the weights of each point data of the humidity distribution set. The temperature distribution fusion weighting process is consistent with the process of fog and humidity. The accuracy of fog, humidity and temperature distribution prediction can be improved by fusion branch construction and branch distribution accuracy testing.

[0035] S30: Analyzing the rationality of the distribution of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model, and analyzing the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, obtaining the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, compensating the fog distribution, humidity distribution and temperature distribution, and obtaining compensated fog distribution, compensated humidity distribution and compensated temperature distribution;

[0036] Further, the distribution rationality of the fog distribution, humidity distribution and temperature distribution is analyzed according to the three-dimensional terrain model, and the detection reproducibility of the fog distribution, humidity distribution and temperature distribution is analyzed according to the historical monitoring data of the measurement scene, the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient are obtained, and the fog distribution, humidity distribution and temperature distribution are compensated. Step S30 includes:

[0037] S31: according to the environmental monitoring data in the multiple measurement scenes, a sample terrain three-dimensional model set, a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set are collected, and a sample fog rationality set, a sample humidity rationality set and a sample temperature rationality set are obtained by evaluation;

[0038] S32: using the sample terrain three-dimensional model set in combination with the sample fog distribution set, the sample humidity distribution set and the sample temperature distribution set as input, respectively using the sample fog rationality set, the sample humidity rationality set and the sample temperature rationality set as output, respectively, to train a fog rationality analysis branch, a humidity rationality analysis branch and a temperature rationality analysis branch;

[0039] S33: using the fog rationality analysis branch, humidity rationality analysis branch and temperature rationality analysis branch, and analyzing and obtaining the fog rationality coefficient, humidity rationality coefficient and temperature rationality coefficient of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model;

[0040] S34: Analyze the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, and obtain the fog reproducibility coefficient, humidity reproducibility coefficient and temperature reproducibility coefficient;

[0041] S35: According to the fog reproducibility coefficient, the humidity reproducibility coefficient and the temperature reproducibility coefficient, the fog rationality coefficient, the humidity rationality coefficient and the temperature rationality coefficient are respectively corrected to obtain the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient;

[0042] S36: Compensating the fog distribution, humidity distribution and temperature distribution according to the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient to obtain compensated fog distribution, compensated humidity distribution and compensated temperature distribution.

[0043] Further, the detection reproducibility of the fog distribution, humidity distribution and temperature distribution is analyzed according to the historical monitoring data of the measurement scene, and step S34 includes:

[0044] S341: using the fog distribution, humidity distribution and temperature distribution, searching in the historical monitoring data of the measurement scene to obtain fog distribution reproducibility parameters, humidity distribution reproducibility parameters and temperature distribution reproducibility parameters;

[0045] S342: Calculate and obtain a fog reproducibility coefficient, a humidity reproducibility coefficient and a temperature reproducibility coefficient according to the fog distribution reproducibility parameter, the humidity distribution reproducibility parameter, the temperature distribution reproducibility parameter and the preset fog reproducibility parameter, the preset humidity reproducibility parameter and the preset temperature reproducibility parameter.

[0046] Specifically, the multiple measurement scenes are different scenes for data collection using radar, and the same terrain data is extracted from the environmental monitoring data of the multiple measurement scenes according to the sample terrain three-dimensional model. If the sample terrain three-dimensional model shows a valley, the valley scene is extracted from several terrain scenes such as the valley scene, the plain scene, and the cliff scene for subsequent data analysis.

[0047] Taking the sample terrain three-dimensional model set as a constraint, a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set are extracted from the environmental monitoring data. The rationality of the sample fog distribution set, the sample humidity distribution set and the sample temperature distribution set is evaluated according to the sample terrain. The fog distribution, humidity distribution and temperature distribution of different terrains have maximum and minimum value restrictions. The rationality of the above distribution sets can be analyzed based on the constraints of terrain attributes combined with empirical judgment to obtain the sample fog rationality set, sample humidity rationality set and sample temperature rationality set.

[0048] The fog rationality analysis branch, humidity rationality analysis branch and temperature rationality analysis branch are branches for judging the rationality of fog distribution, humidity distribution and temperature distribution obtained through a large amount of data training. The sample terrain three-dimensional model set and the sample fog distribution set are used as input data, and the sample fog rationality set is used as output data to train the fog rationality analysis branch. The sample terrain three-dimensional model and the sample humidity distribution set are used as input data, and the sample humidity rationality set is used as output data to train the humidity rationality analysis branch. The sample terrain three-dimensional model and the sample temperature distribution set are used as input data, and the sample temperature rationality set is used as output data to train the temperature rationality analysis branch. After the branch training is completed, the terrain three-dimensional model and the fog distribution are input into the fog rationality analysis branch, the terrain three-dimensional model and the humidity distribution are input into the humidity rationality analysis branch, and the terrain three-dimensional model and the temperature distribution are input into the temperature rationality analysis branch to obtain the rationality coefficients of the three branch outputs, namely the fog rationality coefficient, the humidity rationality coefficient and the temperature rationality coefficient.

[0049] The detection reproducibility indicates the probability of the detection data recurring. The detection reproducibility is achieved by searching in historical data, using fog distribution, humidity distribution and temperature distribution as search conditions, to determine whether the current scene has appeared in the past detection data and how often it appeared. According to the data characteristics and needs, a search method is selected, such as sequential search or binary search. After the data matching judgment is made, the matching successful data is recorded. After traversing all the data, the ratio of the matching successful data to all the data is used as the fog distribution reproducibility parameter, humidity distribution reproducibility parameter and temperature distribution reproducibility parameter.

[0050] Taking the fog reproducibility parameter as an example, the preset reproducibility parameter of fog is explained. The preset reproducibility parameter of fog is the average probability of a fog distribution repeating in all fog monitoring data in the historical data in the scene, such as 10%. Assuming that the probability of the fog distribution reproducibility parameter of the current fog distribution retrieval result is 8%, then the fog reproducibility coefficient is 80%. The larger the reproducibility parameter, the more accurate the current fog distribution. Therefore, the accuracy of the fog distribution, humidity distribution and temperature distribution can be judged by the fog reproducibility coefficient, humidity reproducibility coefficient and temperature reproducibility coefficient. The fog accuracy coefficient is obtained by correcting the fog rationality coefficient according to the fog reproducibility coefficient. For example, it can be obtained by multiplying the fog rationality coefficient by the fog reproducibility coefficient. The humidity accuracy coefficient is obtained by correcting the humidity rationality coefficient according to the humidity reproducibility coefficient, and can be obtained by multiplying the coefficients. The method for obtaining the temperature accuracy coefficient is the same as the previous two. The obtained fog accuracy coefficient, temperature accuracy coefficient, and humidity accuracy coefficient show the deviation between the predicted distribution result and the actual result. The fog distribution, humidity distribution, and temperature distribution are compensated according to the deviation. If the predicted result is lower than the actual result, the distribution after compensation is higher than that before compensation by a certain value. After corresponding compensation through the accuracy coefficient, the compensated fog distribution, compensated humidity distribution, and compensated temperature distribution are obtained, and further correction and adjustment of the distribution data are achieved, so that the fog distribution, humidity distribution, and temperature distribution data are more scientific and accurate.

[0051] S40: collecting radar data collected by radar measurement in the measurement scene, and performing radar data error analysis and correction according to the three-dimensional terrain model to obtain first corrected radar data;

[0052] Further, the radar data collected by radar measurement in the measurement scene is collected, and according to the three-dimensional terrain model, radar data error analysis and correction are performed to obtain first corrected radar data. Step S40 includes:

[0053] S41: Collecting radar data collected by radar measurement in the measurement scene;

[0054] S42: Based on the error influence data of radar data in different terrains, a sample radar data set, a sample terrain three-dimensional model set and a sample first corrected radar data are collected to train and obtain a terrain error radar data corrector;

[0055] S43: Using the terrain error radar data corrector, perform error analysis and correction on the radar data to obtain first corrected radar data.

[0056] Specifically, different terrains have a certain impact on the accuracy of radar data, mainly including terrain blocking. Complex terrains such as mountainous areas may block the radar beam, resulting in missing or inaccurate data in some areas; atmospheric refraction. The atmospheric conditions of different terrains are different, which will affect the refraction of radar waves, thereby causing deviations in the detected target position; altitude differences. The altitudes of different terrains are different, which will affect the radar detection range and echo intensity; beam propagation path differences. In complex terrain environments, the actual propagation path of the radar beam will change. Radar data errors caused by actual terrain require corresponding error analysis and correction. If the data is missing or inaccurate, it needs to be filled and corrected accordingly. False echoes must be identified and corrected.

[0057] The radar data error is corrected by the terrain error radar data corrector. Since the terrain in the measurement scene has a certain degree of complexity and may contain multiple terrains, the three-dimensional models corresponding to the included terrains constitute the sample terrain three-dimensional model set. According to the radar measurement data, the error impact data in different terrains are extracted. For example, complex terrains such as mountainous areas may block the radar beam, resulting in missing or inaccurate data in some areas, which need to be filled and corrected accordingly. The error filled and corrected part belongs to the error impact data. The terrain error radar data corrector is obtained by training based on the neural network model structure. The sample radar data set, the sample terrain three-dimensional model set and the sample first corrected radar data are collected in the error impact data to train the terrain error radar data corrector. The terrain error radar data corrector is used to correct the radar data collected by radar measurement in the measurement scene to obtain the first corrected radar data. The various influences brought by terrain factors are fully considered, and targeted adjustments and corrections are made to improve the accuracy and reliability of the data.

[0058] S50: According to the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, according to the compensated fog distribution, the compensated humidity distribution and the compensated temperature distribution, perform radar data error analysis and correction to obtain second corrected radar data, and fuse with the first corrected radar data to obtain corrected radar data.

[0059] Further, according to the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, according to the compensated fog distribution, the compensated humidity distribution and the compensated temperature distribution, radar data error analysis correction is performed, step S50 includes:

[0060] S51: collecting a sample fog distribution set, a sample radar data set, and a sample fog-corrected radar data set according to error impact data of radar data in different fog distributions;

[0061] S52: using the sample fog distribution set, the sample radar data set and the sample fog correction radar data set, based on ensemble learning, constructing multiple fog radar data correction branches to obtain a fog radar data corrector;

[0062] S53: determining a correction ratio of fog radar data according to the fog accuracy coefficient, randomly selecting a corresponding number of fog radar data correction branches in the fog radar data corrector, performing radar data error analysis and correction on the compensated fog distribution and radar data, obtaining a plurality of fog-corrected radar data, and calculating and obtaining the fog-corrected radar data;

[0063] S54: performing radar data error analysis and correction according to the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, and according to the compensated humidity distribution and the compensated temperature distribution, to obtain humidity-corrected radar data and temperature-corrected radar data;

[0064] S55: fusing the fog-corrected radar data, the humidity-corrected radar data, and the temperature-corrected radar data to obtain second-corrected radar data;

[0065] S56: Perform fusion calculation on the first corrected radar data and the second corrected radar data to obtain corrected radar data.

[0066] Further, the sample fog distribution set, the sample radar data set and the sample fog correction radar data set are used to construct multiple fog radar data correction branches based on ensemble learning, and step S52 includes:

[0067] S521: randomly selecting a preset proportion of sample data from the sample fog distribution set, the sample radar data set, and the sample fog correction radar data set, and constructing and training a first fog radar data correction branch based on ensemble learning;

[0068] S522: Continue to randomly select a preset proportion of sample data from the sample fog distribution set, the sample radar data set, and the sample fog correction radar data set, and construct and train a second fog radar data correction branch based on ensemble learning;

[0069] S523: Based on ensemble learning, continue to construct and obtain multiple other fog radar data correction branches.

[0070] Specifically, different fog distributions will have various effects on radar data, such as weakening radar signal strength, increasing clutter in radar images, reducing resolution, causing signal distortion and errors, etc. Sample data information of the same type of radar under different fog conditions is collected, including a sample fog distribution set, a sample radar data set, and a sample fog-corrected radar data set. The sample fog distribution set, the sample radar data set, and the sample fog-corrected radar data set correspond one to one.

[0071] The fog radar data corrector comprises a plurality of fog radar data correction branches for correcting radar data, wherein the plurality of fog radar data correction branches can process different fog distributions and have different fog distribution pertinences.

[0072] The sample fog distribution set, the sample radar data set and the sample fog correction radar data set are combined as training data, and the first fog radar data correction branch, the second fog radar data correction branch and the Nth fog radar data correction branch are trained in sequence. Exemplarily, the sample data can be classified according to different fog states (such as radiation fog, advection fog, evaporation fog, uphill fog, frontal fog), and the sample data can be divided, such as randomly selecting a preset proportion of sample data, and training different radar data correction branches according to the divided sample data of different categories.

[0073] Ensemble learning can be applied to data correction and error elimination. By combining the prediction results of multiple models, the accuracy and reliability of the data can be improved. Each branch is obtained through ensemble learning. Each branch may include several data correction models, and a fog radar correction branch is formed through ensemble learning. Multiple correction branches are merged, that is, multiple correction branches constitute a fog radar data corrector. By sampling and weighting the training data of different fog radar data correction branches, the first fog radar data correction branch, the second fog radar data correction branch, and the Nth fog radar data correction branch are trained respectively, and their prediction results are processed, such as simple averaging, majority voting, etc., to obtain the final integrated result, that is, multiple fog radar data correction branches are obtained to constitute the fog radar data corrector.

[0074] The fog accuracy coefficient is obtained by correcting the fog rationality coefficient according to the fog reproducibility coefficient. The larger the fog accuracy coefficient, the higher the accuracy of the corresponding compensated fog distribution, and the greater the reliability of the radar data error analysis correction. The fog radar data correction ratio is determined according to the fog accuracy coefficient. For example, if the fog accuracy coefficient is 80%, the number of fog radar data correction branches for radar data error analysis correction is calculated in combination with the number of fog radar data correction branches in the fog radar data corrector. For example, if the number of fog radar data correction branches in the fog radar data corrector is 20, the number of fog radar data correction branches used for radar data correction is 16.

[0075] The number of fog radar data correction branches to be corrected is obtained by multiplying the fog accuracy coefficient by the number of fog radar data correction branches in the fog radar data corrector. The larger the fog accuracy coefficient, the more fog radar data correction branches to be corrected, and the more accurate the result. If the fog accuracy coefficient is small, fewer fog radar data correction branches are used to perform radar data error analysis correction to save processing power for data with lower accuracy, and other types of data are supported to perform radar data error analysis correction, which can ensure accuracy.

[0076] The fog radar data corrector includes multiple fog radar data correction branches. According to the number of fog radar data correction branches determined by the correction ratio, error analysis correction of fog radar data is performed. Specifically, radar data is corrected according to the compensated fog distribution and the fog radar data correction branch input, and multiple fog-corrected radar data can be output. The compensated fog distribution is fog distribution data with higher accuracy. According to the multiple fog-corrected radar data, the fog-corrected radar data is obtained through simple averaging, majority voting, etc., so as to improve the accuracy of fog radar data correction based on integrated learning and save data processing computing power.

[0077] The acquisition method of the humidity-corrected radar data and the temperature-corrected radar data is consistent with the method of the fog-corrected radar data, and the humidity-corrected radar data and the temperature-corrected radar data are obtained through radar data error analysis and correction. The second corrected radar data is obtained by fusing the fog-corrected radar data, the humidity-corrected radar data, and the temperature-corrected radar data, and can be obtained by artificially involving weights and weighted calculation to obtain the second corrected radar data. Furthermore, the first corrected radar data and the second corrected radar data are fused, and after artificially assigning weights, the corrected radar data is obtained by weighted calculation. The secondary correction of the radar collected data is realized, and the technical effect of improving the accuracy of the radar collected data in bad weather is achieved.

[0078] Further, according to the historical monitoring data of the measurement scene, a fog fusion branch, a humidity fusion branch and a temperature fusion branch are constructed, and fusion accuracy analysis is performed to obtain a fog fusion accuracy coefficient set, a humidity fusion accuracy coefficient set and a temperature fusion accuracy coefficient set. Step S22 includes:

[0079] S221: Collecting a sample fog data set, a sample humidity data set and a sample temperature data set, and constructing a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set;

[0080] S222: using the sample fog data set as input, and using the sample fog distribution set, the sample humidity distribution set, and the sample temperature distribution set as output respectively, training the first fog distribution construction path, the first humidity distribution construction path, and the first temperature distribution construction path to obtain fog fusion branches;

[0081] S223: Continue to construct and train to obtain a humidity fusion branch and a temperature fusion branch;

[0082] S224: testing the fog fusion branch to obtain a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient, and a first temperature distribution accuracy coefficient as a fog fusion accuracy coefficient set;

[0083] S225: Testing the humidity fusion branch and the temperature fusion branch to obtain the humidity fusion accuracy coefficient set and the temperature fusion accuracy coefficient set.

[0084] Specifically, the sample fog data set, sample humidity data set and sample temperature data set are extracted from the historical monitoring data of the measurement scene. In order to ensure the validity of the data, data within a certain time threshold from the training time can be selected in the historical monitoring data. Since there will be several fog data, humidity data and temperature data at the same monitoring time, the sample fog distribution set, sample humidity distribution set and sample temperature distribution set are constructed based on these data according to the terrain information of the measurement scene. For example, the distribution of temperature data is constructed based on different temperature data at different coordinate locations in the measurement scene.

[0085] The mist fusion branch includes a first mist distribution construction path, a first humidity distribution construction path and a first temperature distribution construction path, which are parallel models. The first mist distribution construction path, the first humidity distribution construction path and the first temperature distribution construction path are constructed based on a feedforward neural network model. The sample mist data set, the sample humidity data set and the sample temperature data set, as well as the sample mist distribution set, the sample humidity distribution set and the sample temperature distribution set are divided into 8:2, that is, training data: verification data = 8:2. The model is trained using the prepared training data, and the parameters are continuously adjusted to optimize the model performance. The model is evaluated and verified by the verification set, and the model optimization adjustment is further improved according to the verification results of the verification set. The humidity fusion branch and the temperature fusion branch are similar to the mist fusion branch, and are both composed of three parallel models, wherein the humidity fusion branch includes a second mist distribution construction path, a second humidity distribution construction path and a second temperature distribution construction path, and the temperature fusion branch includes a third mist distribution construction path, a third humidity distribution construction path and a third temperature distribution construction path. The training process is similar to that of the mist fusion branch, and no further description is given here.

[0086] The fog fusion branch, humidity fusion branch and temperature fusion branch are tested, and the fog fusion accuracy coefficient set includes the first fog distribution accuracy coefficient, the first humidity distribution accuracy coefficient and the first temperature distribution accuracy coefficient. The humidity fusion accuracy coefficient set includes the second fog distribution accuracy coefficient, the second humidity distribution accuracy coefficient and the second temperature distribution accuracy coefficient. The temperature fusion accuracy coefficient set includes the third fog distribution accuracy coefficient, the third humidity distribution accuracy coefficient and the third temperature distribution accuracy coefficient. The judgment can be made based on the difference between the predicted distribution data and the historical distribution data output during the fusion branch training process. For example, if the temperature distribution difference is 20%, the temperature distribution accuracy coefficient is 80%, and the sum of the two is 100%. For example, by comparing the historical data in the historical sample fog distribution set with the fog distribution output results of the first fog distribution construction path, the second fog distribution construction path, and the third fog distribution construction path, the distribution difference results of different construction paths can be obtained. According to the correlation between accuracy and difference, the accuracy coefficient of each path is obtained. The accuracy coefficient is the mean of the accuracy coefficient test results of each path, including the first fog distribution accuracy coefficient, the first humidity distribution accuracy coefficient, and the first temperature distribution accuracy coefficient. The three accuracy coefficients constitute the fog fusion accuracy coefficient set. Similar to the fog fusion branch test method, the humidity fusion branch and the temperature fusion branch are tested to obtain the humidity fusion accuracy coefficient set and the temperature fusion accuracy coefficient set. Fusion accuracy analysis can provide data support for the subsequent use of distribution data data weighting.

[0087] Example 2

[0088] like Figure 2 As shown, based on the same inventive concept of the radar data collection method combined with scene error analysis provided in Example 1, an embodiment of the present invention provides a radar data collection system combined with scene error analysis, including:

[0089] The terrain data acquisition module 1000 is used to acquire terrain feature data within the measurement scene and construct a three-dimensional terrain model;

[0090] The data fusion module 2000 is used to collect fog data, humidity data, and temperature data in the measurement scene, and perform fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to the historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution;

[0091] The distribution compensation module 3000 is used to analyze the distribution rationality of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model, and to analyze the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, to obtain the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient, to compensate the fog distribution, humidity distribution and temperature distribution, and to obtain the compensated fog distribution, the compensated humidity distribution and the compensated temperature distribution;

[0092] The first correction module 4000 is used to collect radar data collected by radar measurement in the measurement scene, perform radar data error analysis and correction according to the three-dimensional terrain model, and obtain first corrected radar data;

[0093] The second correction module 5000 is used to perform radar data error analysis and correction based on the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, based on integrated learning, and according to the compensated fog distribution, compensated humidity distribution and compensated temperature distribution, to obtain second corrected radar data, and fuse it with the first corrected radar data to obtain corrected radar data.

[0094] Furthermore, the system also includes:

[0095] A marking point data collection module, used to collect fog data, humidity data, and temperature data of multiple marking points in the measurement scene;

[0096] a fusion analysis module, configured to construct a fog fusion branch, a humidity fusion branch, and a temperature fusion branch according to the historical monitoring data of the measurement scene, and perform fusion accuracy analysis to obtain a fog fusion accuracy coefficient set, a humidity fusion accuracy coefficient set, and a temperature fusion accuracy coefficient set, wherein the fog fusion branch includes a first fog distribution construction path, a first humidity distribution construction path, and a first temperature distribution construction path, and the fog fusion accuracy coefficient set includes a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient, and a first temperature distribution accuracy coefficient;

[0097] A branch processing module, used for inputting the fog data, humidity data, and temperature data into the fog fusion branch, humidity fusion branch, and temperature fusion branch, respectively, and processing to obtain a fog distribution set, a humidity distribution set, and a temperature distribution set;

[0098] The fusion weighting module is used to use the fog fusion accuracy coefficient set, humidity fusion accuracy coefficient set and temperature fusion accuracy coefficient set to perform fusion weighted processing on the fog distribution set, humidity distribution set and temperature distribution set to obtain fog distribution, humidity distribution and temperature distribution.

[0099] Furthermore, the system also includes:

[0100] A sample data collection module is used to collect a sample fog data set, a sample humidity data set and a sample temperature data set, and construct a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set;

[0101] a fusion branch training module, configured to use the sample fog data set as input, and respectively use the sample fog distribution set, the sample humidity distribution set, and the sample temperature distribution set as output, to train the first fog distribution construction path, the first humidity distribution construction path, and the first temperature distribution construction path, to obtain a fog fusion branch;

[0102] A branch construction module is used to continue to construct training to obtain humidity fusion branches and temperature fusion branches;

[0103] A fog branch testing module, used for testing the fog fusion branch to obtain a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient and a first temperature distribution accuracy coefficient as a fog fusion accuracy coefficient set;

[0104] The temperature and humidity branch testing module is used to test the humidity fusion branch and the temperature fusion branch to obtain the humidity fusion accuracy coefficient set and the temperature fusion accuracy coefficient set.

[0105] Furthermore, the system also includes:

[0106] A rationality evaluation module is used to collect a sample terrain three-dimensional model set, a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set based on environmental monitoring data in multiple measurement scenes, and evaluate and obtain a sample fog rationality set, a sample humidity rationality set and a sample temperature rationality set;

[0107] A rationality analysis branch construction module is used to use the sample terrain three-dimensional model set in combination with the sample fog distribution set, the sample humidity distribution set and the sample temperature distribution set as input, and use the sample fog rationality set, the sample humidity rationality set and the sample temperature rationality set as output to train the fog rationality analysis branch, the humidity rationality analysis branch and the temperature rationality analysis branch;

[0108] A rationality coefficient analysis module, for adopting the fog rationality analysis branch, the humidity rationality analysis branch and the temperature rationality analysis branch, and analyzing and obtaining the fog rationality coefficient, the humidity rationality coefficient and the temperature rationality coefficient of the fog distribution, the humidity distribution and the temperature distribution according to the three-dimensional terrain model;

[0109] A reproducibility analysis module, for analyzing the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, and obtaining a fog reproducibility coefficient, a humidity reproducibility coefficient and a temperature reproducibility coefficient;

[0110] A coefficient correction module, for correcting the fog rationality coefficient, humidity rationality coefficient, and temperature rationality coefficient respectively according to the fog reproducibility coefficient, humidity reproducibility coefficient, and temperature reproducibility coefficient, to obtain a fog accuracy coefficient, a temperature accuracy coefficient, and a humidity accuracy coefficient;

[0111] The accurate distribution compensation module is used to compensate the fog distribution, humidity distribution and temperature distribution according to the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient to obtain compensated fog distribution, compensated humidity distribution and compensated temperature distribution.

[0112] Furthermore, the system also includes:

[0113] A data retrieval module, for using the fog distribution, humidity distribution and temperature distribution to search in the historical monitoring data of the measurement scene to obtain the fog distribution reproducibility parameter, the humidity distribution reproducibility parameter and the temperature distribution reproducibility parameter;

[0114] The reproducibility coefficient analysis module is used to calculate the fog reproducibility coefficient, humidity reproducibility coefficient and temperature reproducibility coefficient according to the fog distribution reproducibility parameter, humidity distribution reproducibility parameter, temperature distribution reproducibility parameter and the preset fog reproducibility parameter, preset humidity reproducibility parameter and preset temperature reproducibility parameter.

[0115] Furthermore, the system also includes:

[0116] A first radar data acquisition module, used to acquire radar data for radar measurement acquisition in the measurement scene;

[0117] A terrain data correction module is used to collect a sample radar data set, a sample terrain three-dimensional model set and a sample first corrected radar data based on the error impact data of radar data in different terrains, and train to obtain a terrain error radar data corrector;

[0118] The first error analysis and correction module is used to use the terrain error radar data corrector to perform error analysis and correction on the radar data to obtain first corrected radar data.

[0119] Furthermore, the system also includes:

[0120] The second radar data collection module is used to collect a sample fog distribution set, a sample radar data set and a sample fog-corrected radar data set according to the error impact data of the radar data in different fog distributions;

[0121] A fog data correction module, configured to adopt the sample fog distribution set, the sample radar data set and the sample fog correction radar data set, construct a plurality of fog radar data correction branches based on ensemble learning, and obtain a fog radar data corrector;

[0122] A fog correction data calculation module is used to determine the fog radar data correction ratio according to the fog accuracy coefficient, randomly select a corresponding number of fog radar data correction branches in the fog radar data corrector, perform radar data error analysis and correction on the compensated fog distribution and radar data, obtain multiple fog corrected radar data, and calculate and obtain the fog corrected radar data;

[0123] A temperature and humidity correction module, for performing radar data error analysis and correction based on the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, and based on the compensated humidity distribution and the compensated temperature distribution, to obtain humidity-corrected radar data and temperature-corrected radar data;

[0124] A correction data fusion module, used for fusing the fog correction radar data, the humidity correction radar data and the temperature correction radar data to obtain second correction radar data;

[0125] The data fusion analysis module is used to perform fusion calculation on the first corrected radar data and the second corrected radar data to obtain corrected radar data.

[0126] Furthermore, the system also includes:

[0127] A first branch construction module is used to randomly select a preset proportion of sample data from the sample fog distribution set, the sample radar data set and the sample fog correction radar data set, and to construct and train a first fog radar data correction branch based on ensemble learning;

[0128] A second branch construction module is used to continue to randomly select a preset proportion of sample data from the sample fog distribution set, the sample radar data set and the sample fog correction radar data set, and to construct and train a second fog radar data correction branch based on ensemble learning;

[0129] The third branch construction module is used to continue to construct and obtain other multiple fog radar data correction branches based on integrated learning.

[0130] See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention proposes an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, the following steps are implemented: collecting terrain feature data in a measurement scene to construct a three-dimensional terrain model; collecting fog data, humidity data, and temperature data in the measurement scene, and performing fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution; analyzing the distribution rationality of the fog distribution, humidity distribution, and temperature distribution according to the three-dimensional terrain model, and, according to the historical monitoring data of the measurement scene, analyzing the distribution rationality of the fog distribution, humidity distribution, and temperature distribution. Analyze the detection reproducibility of the fog distribution, humidity distribution and temperature distribution, obtain the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, compensate the fog distribution, humidity distribution and temperature distribution, and obtain compensated fog distribution, compensated humidity distribution and compensated temperature distribution; collect radar data collected by radar measurement in the measurement scene, perform radar data error analysis and correction according to the terrain three-dimensional model, and obtain first corrected radar data; according to the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, based on ensemble learning, perform radar data error analysis and correction according to the compensated fog distribution, compensated humidity distribution and compensated temperature distribution, obtain second corrected radar data, and fuse them with the first corrected radar data to obtain corrected radar data.

[0131] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0132] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0134] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0135] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0136] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0137] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technology, the present invention is also intended to include these changes and variations.

Claims

1. A radar data acquisition method combined with scene error analysis, characterized in that: The method comprises: Collect terrain feature data within the measurement scene and construct a three-dimensional terrain model; Collect fog data, humidity data, and temperature data in the measurement scene, and perform fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution; Analyzing the rationality of the distribution of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model, and analyzing the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, obtaining the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, compensating the fog distribution, humidity distribution and temperature distribution, and obtaining compensated fog distribution, compensated humidity distribution and compensated temperature distribution; Collecting radar data collected by radar measurement in the measurement scene, and performing radar data error analysis and correction according to the three-dimensional terrain model to obtain first corrected radar data; According to the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, based on integrated learning, according to the compensated fog distribution, compensated humidity distribution and compensated temperature distribution, radar data error analysis and correction are performed to obtain second corrected radar data, which are then fused with the first corrected radar data to obtain corrected radar data.

2. The method according to claim 1, characterized in that Collect fog data, humidity data, and temperature data in the measurement scene, and perform fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to historical monitoring data of the measurement scene, including: Collect fog data, humidity data, and temperature data of multiple marked points in the measurement scene; According to the historical monitoring data of the measurement scene, a fog fusion branch, a humidity fusion branch and a temperature fusion branch are constructed, and fusion accuracy analysis is performed to obtain a fog fusion accuracy coefficient set, a humidity fusion accuracy coefficient set and a temperature fusion accuracy coefficient set, wherein the fog fusion branch includes a first fog distribution construction path, a first humidity distribution construction path and a first temperature distribution construction path, and the fog fusion accuracy coefficient set includes a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient and a first temperature distribution accuracy coefficient; The fog data, humidity data, and temperature data are input into the fog fusion branch, the humidity fusion branch, and the temperature fusion branch respectively, and processed to obtain a fog distribution set, a humidity distribution set, and a temperature distribution set; The fog fusion accuracy coefficient set, humidity fusion accuracy coefficient set and temperature fusion accuracy coefficient set are used to perform fusion weighted processing on the fog distribution set, humidity distribution set and temperature distribution set to obtain fog distribution, humidity distribution and temperature distribution.

3. The method according to claim 2, characterized in that According to the historical monitoring data of the measurement scene, a fog fusion branch, a humidity fusion branch and a temperature fusion branch are constructed, and a fusion accuracy analysis is performed to obtain a fog fusion accuracy coefficient set, a humidity fusion accuracy coefficient set and a temperature fusion accuracy coefficient set, including: Collecting a sample fog data set, a sample humidity data set and a sample temperature data set, and constructing a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set; Using the sample fog data set as input, and using the sample fog distribution set, the sample humidity distribution set, and the sample temperature distribution set as output respectively, training the first fog distribution construction path, the first humidity distribution construction path, and the first temperature distribution construction path to obtain fog fusion branches; Continue to build and train to obtain humidity fusion branch and temperature fusion branch; Testing the fog fusion branch to obtain a first fog distribution accuracy coefficient, a first humidity distribution accuracy coefficient, and a first temperature distribution accuracy coefficient as a fog fusion accuracy coefficient set; The humidity fusion branch and the temperature fusion branch are tested to obtain the humidity fusion accuracy coefficient set and the temperature fusion accuracy coefficient set.

4. The method according to claim 2, characterized in that Analyzing the rationality of the distribution of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model, and analyzing the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, obtaining the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient, and compensating the fog distribution, humidity distribution and temperature distribution, including: According to the environmental monitoring data in multiple measurement scenes, a sample terrain three-dimensional model set, a sample fog distribution set, a sample humidity distribution set and a sample temperature distribution set are collected, and the sample fog rationality set, the sample humidity rationality set and the sample temperature rationality set are obtained by evaluation; The sample terrain three-dimensional model set is used in combination with the sample fog distribution set, the sample humidity distribution set and the sample temperature distribution set as input, and the sample fog rationality set, the sample humidity rationality set and the sample temperature rationality set are used as output to train the fog rationality analysis branch, the humidity rationality analysis branch and the temperature rationality analysis branch; The fog rationality analysis branch, humidity rationality analysis branch and temperature rationality analysis branch are used to analyze and obtain the fog rationality coefficient, humidity rationality coefficient and temperature rationality coefficient of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model; Analyzing the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, and obtaining the fog reproducibility coefficient, humidity reproducibility coefficient and temperature reproducibility coefficient; According to the fog reproducibility coefficient, humidity reproducibility coefficient and temperature reproducibility coefficient, the fog rationality coefficient, humidity rationality coefficient and temperature rationality coefficient are calibrated respectively to obtain the fog accuracy coefficient, temperature accuracy coefficient and humidity accuracy coefficient; The fog distribution, humidity distribution and temperature distribution are compensated according to the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient to obtain compensated fog distribution, compensated humidity distribution and compensated temperature distribution.

5. The method according to claim 4, characterized in that Analyze the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, including: Using the fog distribution, humidity distribution and temperature distribution, searching in the historical monitoring data of the measurement scene to obtain fog distribution reproducibility parameters, humidity distribution reproducibility parameters and temperature distribution reproducibility parameters; According to the mist distribution reproducibility parameter, humidity distribution reproducibility parameter, temperature distribution reproducibility parameter and the preset mist reproducibility parameter, preset humidity reproducibility parameter, preset temperature reproducibility parameter, the mist reproducibility coefficient, humidity reproducibility coefficient and temperature reproducibility coefficient are calculated.

6. The method according to claim 1, characterized in that Collecting radar data collected by radar measurement in the measurement scene, and performing radar data error analysis and correction according to the three-dimensional terrain model to obtain first corrected radar data, including: Collecting radar data collected by radar measurement in the measurement scene; Based on the error impact data of radar data in different terrains, a sample radar data set, a sample terrain three-dimensional model set and a sample first corrected radar data are collected to train and obtain a terrain error radar data corrector; The terrain error radar data corrector is used to perform error analysis and correction on the radar data to obtain first corrected radar data.

7. The method according to claim 1, characterized in that According to the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, according to the compensated fog distribution, the compensated humidity distribution and the compensated temperature distribution, radar data error analysis correction is performed, including: According to the error impact data of radar data in different fog distributions, a sample fog distribution set, a sample radar data set and a sample fog-corrected radar data set are collected; Using the sample fog distribution set, the sample radar data set and the sample fog correction radar data set, based on ensemble learning, multiple fog radar data correction branches are constructed to obtain a fog radar data corrector; According to the fog accuracy coefficient, a fog radar data correction ratio is determined, a corresponding number of fog radar data correction branches are randomly selected in the fog radar data corrector, radar data error analysis and correction is performed on the compensated fog distribution and radar data, a plurality of fog corrected radar data are obtained, and the fog corrected radar data is obtained by calculation; According to the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, radar data error analysis and correction are performed according to compensated humidity distribution and compensated temperature distribution to obtain humidity corrected radar data and temperature corrected radar data; fusing the fog-corrected radar data, the humidity-corrected radar data, and the temperature-corrected radar data to obtain second-corrected radar data; The first corrected radar data and the second corrected radar data are fused and calculated to obtain corrected radar data.

8. The method according to claim 7, characterized in that The sample fog distribution set, the sample radar data set and the sample fog correction radar data set are used to construct multiple fog radar data correction branches based on ensemble learning, including: Randomly select a preset proportion of sample data from the sample fog distribution set, the sample radar data set, and the sample fog correction radar data set, and construct and train a first fog radar data correction branch based on ensemble learning; Continue to randomly select a preset proportion of sample data from the sample fog distribution set, the sample radar data set, and the sample fog correction radar data set, and construct and train a second fog radar data correction branch based on ensemble learning; Based on ensemble learning, we continue to build and obtain multiple other fog radar data correction branches.

9. A radar data acquisition system combined with scene error analysis, characterized in that: include: The terrain data acquisition module is used to collect terrain feature data within the measurement scene and build a three-dimensional terrain model; A data fusion module is used to collect fog data, humidity data, and temperature data in the measurement scene, and perform fusion accuracy analysis and fusion processing on the fog data, humidity data, and temperature data according to historical monitoring data of the measurement scene to obtain fog distribution, humidity distribution, and temperature distribution; a distribution compensation module, for analyzing the distribution rationality of the fog distribution, humidity distribution and temperature distribution according to the three-dimensional terrain model, and analyzing the detection reproducibility of the fog distribution, humidity distribution and temperature distribution according to the historical monitoring data of the measurement scene, obtaining the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient, compensating the fog distribution, humidity distribution and temperature distribution, and obtaining compensated fog distribution, compensated humidity distribution and compensated temperature distribution; A first correction module is used to collect radar data collected by radar measurement in the measurement scene, and perform radar data error analysis and correction according to the three-dimensional terrain model to obtain first corrected radar data; The second correction module is used to perform radar data error analysis and correction based on the fog accuracy coefficient, the temperature accuracy coefficient and the humidity accuracy coefficient, based on integrated learning, and according to the compensated fog distribution, the compensated humidity distribution and the compensated temperature distribution, to obtain second corrected radar data, and fuse the second corrected radar data with the first corrected radar data to obtain corrected radar data.

10. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the radar data collection method combined with scene error analysis as described in any one of claims 1 to 8.

Images