Data association detection method and device, computer equipment and storage medium
By denoising and fusion processing of the trajectory data of radar equipment and AIS, trajectory characteristics are determined, and correlation detection is performed using classification models, the problem of low accuracy of correlation state detection results in the prior art is solved, and higher detection accuracy is achieved.
Patent Information
- Application Number
- CN202411864073.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-17
- Publication Date
- 2025-04-18
AI Technical Summary
In the prior art, when performing correlation detection of the trajectory between the radar device and the AIS, the correctness of the correlation state detection results is low.
By obtaining the radar trajectory data and AIS trajectory data of the target vehicle, performing denoising processing and performing fusion processing, trajectory characteristics are determined, and these characteristics are input into the pre-constructed classification model for correlation detection.
The accuracy of the trajectory correlation detection results between radar equipment and AIS is improved, making the dependent data of correlation detection more comprehensive and accurate, and achieving accurate correlation detection.
Smart Images
Figure CN119848620A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of transportation technology, and in particular to a data association detection method, device, computer equipment and storage medium. Background Art
[0002] With the continuous development of science and technology, the transportation industry is also constantly changing. In order to ensure the safety and efficiency of transportation, it is necessary to obtain the precise trajectory of vehicles during driving.
[0003] Taking water vehicles as an example, in actual applications, the trajectory data of water vehicles is detected by radar equipment and the Automatic Identification System (AIS) respectively, and then these trajectory data are associated to obtain the accurate trajectory data of the water vehicle during driving. Therefore, it is particularly important to obtain the association status between the trajectory data detected by the radar equipment and the AIS before the trajectory data is associated.
[0004] In the related art, statistical methods are mainly used to perform correlation detection on the trajectories between the radar device and the AIS based on the trajectory data of the water vehicle detected by the radar device and the AIS, so as to determine the correlation status between different trajectories.
[0005] However, in the related art, in the process of performing association detection on the tracks between the radar device and the AIS, there is a problem that the accuracy of the association status detection result is low. Summary of the invention
[0006] Based on this, it is necessary to provide a data association detection method, device, computer equipment and storage medium to address the above technical problems, which can improve the accuracy of the association status detection results in the process of association detection of the trajectory between the radar equipment and the AIS.
[0007] In a first aspect, an embodiment of the present application provides a data association detection method, the method comprising:
[0008] Obtain radar track data and ship automatic identification system AIS track data of target vehicles;
[0009] De-noising the radar trajectory data to obtain first trajectory data; and De-noising the AIS trajectory data to obtain second trajectory data;
[0010] Determine the trajectory characteristics of the target vehicle by fusing the first trajectory data and the second trajectory data;
[0011] The trajectory features are input into the pre-built classification model to perform correlation detection on the radar trajectory and AIS trajectory of the target vehicle, and the correlation status between the radar trajectory and AIS trajectory of the target vehicle is obtained.
[0012] In one embodiment, obtaining radar track data and automatic ship identification system AIS track data of a target vehicle includes:
[0013] Obtain the original radar track data and original AIS track data of the target vehicle;
[0014] The original radar track data and the original AIS track data are time aligned to obtain radar track data and AIS track data.
[0015] In one embodiment, performing denoising on the radar trajectory data to obtain first trajectory data includes:
[0016] Encoding the radar trajectory data to obtain radar coded trajectory data of the target vehicle;
[0017] The radar coded trajectory data is decoded to obtain first trajectory data.
[0018] In one embodiment, performing denoising on the AIS trajectory data to obtain second trajectory data includes:
[0019] Encoding the AIS trajectory data to obtain the AIS coded trajectory data of the target vehicle;
[0020] The AIS encoded trajectory data is decoded to obtain second trajectory data.
[0021] In one embodiment, the step of fusing the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle includes:
[0022] Performing difference processing on the first track data and the second track data to determine position difference data, speed difference data and heading difference data between the radar track and the AIS track;
[0023] The trajectory characteristics of the target vehicle are constructed based on the position difference data, speed difference data and heading difference data.
[0024] In one embodiment, constructing the trajectory characteristics of the target vehicle according to the position difference data, the speed difference data and the heading difference data includes:
[0025] The position difference data, speed difference data and heading difference data are combined to generate trajectory features.
[0026] In one embodiment, the process of building a classification model includes:
[0027] Obtain trajectory feature sets and trajectory data sets of different vehicles; the trajectory data sets include multiple labeled radar trajectory data and labeled AIS trajectory data;
[0028] According to the trajectory data set and the trajectory data set, the initial classification model is trained to obtain a trained classification model.
[0029] In a second aspect, an embodiment of the present application provides a data association detection device, the device comprising:
[0030] An acquisition module is used to acquire radar track data of target vehicles and ship automatic identification system AIS track data;
[0031] A denoising processing module, configured to perform denoising processing on the radar trajectory data to obtain first trajectory data; and perform denoising processing on the AIS trajectory data to obtain second trajectory data;
[0032] A determination module, configured to perform a fusion process on the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle;
[0033] The association detection module is used to input the trajectory features into a pre-built classification model to perform association detection on the radar trajectory and the AIS trajectory of the target vehicle, and obtain the association status between the radar trajectory and the AIS trajectory of the target vehicle.
[0034] In a third aspect, an embodiment of the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of any one of the embodiments of the first aspect when executing the computer program.
[0035] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method of any embodiment in the above-mentioned first aspect are implemented.
[0036] In a fifth aspect, an embodiment of the present application further provides a computer program product, the computer program product comprising a computer program, which, when executed by a processor, implements the steps of the method of any embodiment of the first aspect above.
[0037] The embodiment of the present application provides a data association detection method, device, computer equipment and storage medium, including: obtaining radar trajectory data and ship automatic identification system AIS trajectory data of a target vehicle, denoising the radar trajectory data to obtain first trajectory data, and denoising the AIS trajectory data to obtain second trajectory data, determining the trajectory characteristics of the target vehicle according to the fusion processing of the first trajectory data and the second trajectory data, inputting the trajectory characteristics into a pre-built classification model to perform association detection on the radar trajectory and AIS trajectory of the target vehicle, and obtaining the association state between the radar trajectory and AIS trajectory of the target vehicle. The above method can be used to denoise the trajectory data corresponding to the target vehicle obtained by different devices, and then the trajectory data with higher purity after denoising can be used to realize association detection, which can improve the accuracy of the association detection result; at the same time, the above method can fuse the denoising trajectory data of the target vehicle corresponding to different devices, so that the dependent data of the association detection is more comprehensive and accurate, so as to prepare for the realization of accurate association detection, and greatly improve the accuracy of the association detection result, and on this basis, it can provide a reliable reference for applications based on trajectory data. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 is an application environment diagram of a data association detection method in an embodiment;
[0039] Figure 2 1 is a flow chart of a data association detection method in one embodiment;
[0040] Figure 3 A schematic diagram of the internal structure of a classification model in one embodiment;
[0041] Figure 4 is a flow chart of a data association detection method in another embodiment;
[0042] Figure 5 is a flow chart of a data association detection method in another embodiment;
[0043] Figure 6 A diagram showing corresponding results of trajectory data before and after time alignment processing in one embodiment;
[0044] Figure 7 is a flow chart of a data association detection method in another embodiment;
[0045] Figure 8 A schematic diagram of the internal structure of a denoising processing module in one embodiment;
[0046] Fig. 9 is a flow chart of a data association detection method in another embodiment;
[0047] Fig.10 is a flow chart of a data association detection method in another embodiment;
[0048] Fig.11 is a structural block diagram of a data association detection device in one embodiment;
[0049] Fig.12 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0050] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0051] In the field of transportation, in order to ensure the safety and efficiency of transportation, it is necessary to obtain the precise trajectory of the vehicle during its travel. Taking water vehicles as an example, in actual applications, the trajectory data of water vehicles is detected by radar equipment and the Automatic Identification System (AIS) respectively, and then these trajectory data are associated to obtain the precise trajectory data of the water vehicle during its travel. Therefore, it is particularly important to obtain the association status between the trajectory data detected by the radar equipment and the AIS before the trajectory data is associated.
[0052] In the related art, statistical methods are mainly used to perform correlation detection on the trajectories between the radar device and the AIS based on the trajectory data of the water vehicle detected by the radar device and the AIS to determine the correlation status between different trajectories. However, in the related art, in the process of performing correlation detection on the trajectories between the radar device and the AIS, there is a problem of low accuracy of the correlation status detection result. Based on this, an embodiment of the present application provides a data correlation detection method, which can improve the accuracy of the correlation status detection result in the process of performing correlation detection on the trajectories between the radar device and the AIS.
[0053] The data association detection method provided in the embodiment of the present application can be applied to Figure 1The application environment diagram shown. The application environment includes a data association detection system, which may include radar equipment, multiple water vehicles, AIS installed on each water vehicle, and computer equipment; the deployment position of the radar equipment is not fixed, as long as it can cover each water vehicle. Among them, the computer equipment can be connected to the radar equipment and the AIS for communication, and the communication method can be Wi-Fi, mobile network or Bluetooth connection, etc. Optionally, the above-mentioned radar equipment can be a millimeter wave radar, a phased array radar, an active electronic scanning array radar; the water vehicle can be a rowing boat, a fishing boat, a ferry, a cruise ship, a ferry, etc., but in the embodiment of the present application, the water vehicle is a ship as an example for explanation. At the same time, the above-mentioned computer equipment can be a personal computer, a desktop computer, a laptop computer, a tablet computer, an iPad, a smart watch, a smart bracelet and other electronic devices. The following embodiment uses a computer device as the execution subject of the data association detection method to introduce the specific process of the data association detection method.
[0054] like Figure 2 FIG. 1 is a flow chart of a data association detection method provided in an embodiment of the present application. The method is applied to a computer device and can be implemented by the following steps:
[0055] S100: Acquire radar track data and Automatic Identification System (AIS) track data of the target vehicle.
[0056] In practical applications, the target vehicle can be any water vehicle; the radar trajectory data and AIS trajectory data of the target vehicle are time-synchronized trajectory data. The radar trajectory data is the corresponding trajectory data collected by the radar device within a period of time, and the AIS trajectory data is the corresponding trajectory data collected by the AIS within the same period of time.
[0057] Specifically, the computer device can read the radar track data and the ship automatic identification system AIS track data of the target vehicle from a local location, a cloud location, a disk, a hard disk, etc. In addition, the computer device can also send a track data acquisition request to a third-party device, and the third-party device responds to the track data acquisition request and feeds back the radar track data and the ship automatic identification system AIS track data of the target vehicle to the computer device. Correspondingly, the computer device receives and acquires the radar track data and the ship automatic identification system AIS track data of the target vehicle.
[0058] S200: performing denoising processing on radar trajectory data to obtain first trajectory data; and performing denoising processing on AIS trajectory data to obtain second trajectory data.
[0059] Among them, the computer device can pre-train an algorithm model, and then input the radar trajectory data into the algorithm model to denoise the radar trajectory data and output the first trajectory data; at the same time, the AIS trajectory data can be input into the algorithm model to denoise the AIS trajectory data and output the second trajectory data.
[0060] In addition, the computer device may first perform missing value processing, outlier replacement, duplicate value deletion and / or standardization processing on the radar trajectory data to obtain processed radar trajectory data, and then use a data denoising processing algorithm to perform denoising processing on the processed radar trajectory data to obtain first trajectory data. At the same time, the computer device may first perform missing value processing, outlier replacement, duplicate value deletion and / or standardization processing on the AIS trajectory data to obtain processed AIS trajectory data, and then use a data denoising processing algorithm to perform denoising processing on the processed AIS trajectory data to obtain second trajectory data.
[0061] S300: Perform fusion processing on the first trajectory data and the second trajectory data to determine trajectory characteristics of the target vehicle.
[0062] In practical applications, the computer device may pre-train a fusion processing model, and then input both the first trajectory data and the second trajectory data into the fusion processing model, and the trajectory feature extraction model outputs the trajectory features of the target vehicle.
[0063] Optionally, the above-mentioned fusion processing model can be implemented by at least one of a convolutional neural network model, a fully connected neural network model, a long short-term memory neural network model, a recurrent neural network model, etc.
[0064] S400: Input the trajectory feature into a pre-built classification model to perform correlation detection on the radar trajectory and the AIS trajectory of the target vehicle, and obtain the correlation state between the radar trajectory and the AIS trajectory of the target vehicle.
[0065] The above classification model can be implemented by at least one of a fully connected neural network model, a long short-term memory neural network model, a residual neural network model, a decision tree, etc. In the embodiment of the present application, the above classification model is constructed by a convolutional neural network model, such as Figure 3 As shown, the classification model includes two convolutional layers, two fully connected layers, an activation layer and an output layer in sequence. In an embodiment of the present application, the above classification model is a supervised learning model.
[0066] Optionally, the association state between the radar track and the AIS track of the target vehicle may be that the radar track and the AIS track are associated or not associated. In practical applications, if the label label output by the output layer in the classification model is 1, it indicates that the radar track and the AIS track of the target vehicle are associated; if the label label output by the output layer in the classification model is 0, it indicates that the radar track and the AIS track of the target vehicle are not associated.
[0067] In one embodiment, if Figure 4 As shown, the construction process of the above classification model may include:
[0068] S500: Acquire trajectory feature sets and trajectory data sets of different vehicles, wherein the trajectory data sets include a plurality of labeled radar trajectory data and labeled AIS trajectory data.
[0069] In practical applications, the above-mentioned labeled radar trajectory data may include labeled radar trajectory data after denoising and labeled radar trajectory data before denoising; the above-mentioned labeled AIS trajectory data may include labeled AIS trajectory data after denoising and labeled AIS trajectory data before denoising. The vehicle in S500 above refers to a water vehicle, which is the same type as the target vehicle.
[0070] The computer device may determine the trajectory feature sets of different vehicles for all denoised radar trajectory data and all denoised AIS trajectory data corresponding to different vehicles.
[0071] S600 , training an initial classification model according to the trajectory data set and the trajectory data set to obtain a trained classification model.
[0072] Specifically, the computer device can input both the trajectory data set and the trajectory feature set into the initial classification model to obtain the predicted association state between the radar trajectory and the AIS trajectory of different vehicles, calculate the prediction error value between the predicted association state and the standard association state through the loss function, and update the initial network parameters in the initial classification model according to the prediction error value, and continuously iterate the above training steps until the prediction error value meets the preset error threshold or the number of iterations reaches the preset iteration number threshold, so as to obtain the classification model. The above standard association state can be an idealized association state.
[0073] Optionally, the above loss function can be at least one of a mean square error loss function, a mean absolute error loss function, a cross entropy loss function, an exponential loss function, an absolute value loss function, and the like.
[0074] The technical solution in the embodiment of the present application obtains radar trajectory data and ship automatic identification system AIS trajectory data of the target vehicle, performs denoising processing on the radar trajectory data to obtain first trajectory data, and performs denoising processing on the AIS trajectory data to obtain second trajectory data, performs fusion processing on the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle, and inputs the trajectory characteristics into a pre-built classification model to perform association detection on the radar trajectory and the AIS trajectory of the target vehicle to obtain the association state between the radar trajectory and the AIS trajectory of the target vehicle; the above method can perform denoising processing on the trajectory data corresponding to the target vehicle obtained by different devices, and then perform association detection on the trajectory data after denoising with higher purity, which can improve the accuracy of the association detection result; at the same time, the above method can perform fusion processing on the denoised trajectory data of the target vehicle corresponding to different devices, so that the dependent data of the association detection is more comprehensive and accurate, so as to prepare for the realization of accurate association detection, and greatly improve the accuracy of the association detection result, and on this basis, it can provide a reliable reference basis for applications based on trajectory data.
[0075] The following is an explanation of the process of obtaining the radar trajectory data of the target vehicle and the AIS trajectory data of the ship automatic identification system. In one embodiment, the steps in S100 are as follows: Figure 5 As shown, this can be achieved by:
[0076] S110: Acquire original radar track data and original AIS track data of the target vehicle.
[0077] The computer device may obtain the original radar trajectory data of the target vehicle directly collected by the radar device and the original AIS trajectory data of the target vehicle directly collected by the AIS installed on the target vehicle.
[0078] S120 , performing time alignment processing on the original radar trajectory data and the original AIS trajectory data to obtain radar trajectory data and AIS trajectory data.
[0079] Specifically, the computer device may pre-train a time alignment model, and then input both the original radar trajectory data and the original AIS trajectory data into the time alignment model, which performs time alignment processing on the original radar trajectory data and the original AIS trajectory data and then outputs the radar trajectory data and the AIS trajectory data.
[0080] Optionally, the above-mentioned time alignment model can be implemented by at least one of a fully connected neural network model, a long short-term memory neural network model, a residual neural network model, etc.
[0081] In an embodiment of the present application, the original radar trajectory data may include trajectory positions, trajectory speeds, and trajectory headings corresponding to multiple moments; at the same time, the original AIS trajectory data may also include trajectory positions, trajectory speeds, and trajectory headings corresponding to multiple moments. Specifically, the computer device may use the Lagrange interpolation method to perform time alignment processing on the original radar trajectory data and the original AIS trajectory data. Optionally, the above-mentioned Lagrange interpolation method may be a Lagrange two-point interpolation method, a Lagrange four-point interpolation method, etc. In an embodiment of the present application, the Lagrange interpolation method is a Lagrange three-point interpolation method as an example for explanation.
[0082] In practical applications, after the original radar trajectory data and the original AIS trajectory data are time-aligned, the acquired radar trajectory data and AIS trajectory data both include k trajectory positions corresponding to the same time. The following describes the process of time-aligning the original radar trajectory data and the original AIS trajectory data using the Lagrange three-point interpolation method in the embodiment of the present application:
[0083] For the original radar track data or the original AIS track data, the time interval of each three adjacent moments ( , and ) correspond to the three trajectory positions, namely , , , and then according to these three moments, three trajectory positions and the preset interpolation moment To calculate The trajectory position corresponding to the time , respectively expressed as:
[0084] + (1)
[0085] + (2)
[0086] For example, Figure 6 As shown in the figure, in order to time align the original radar trajectory data and the original AIS trajectory data, the trajectory positions corresponding to 1s, 3s, 5s and 7s are inserted into the original radar trajectory data. At the same time, the trajectory positions corresponding to 1s, 2s, 4s, 5s, 7s and 8s are inserted into the original AIS trajectory data. The finally acquired radar trajectory data and AIS trajectory data both include the trajectory positions corresponding to 10 moments.
[0087] The technical solution in the embodiment of the present application can obtain the original radar trajectory data and the original AIS trajectory data of the target vehicle, and perform time alignment processing on the original radar trajectory data and the original AIS trajectory data to obtain radar trajectory data and AIS trajectory data, so that the acquired radar trajectory data and AIS trajectory data are smoother in time, so as to provide reliable dependency information for subsequent precise processing.
[0088] The following is an explanation of the process of performing denoising on the radar trajectory data to obtain the first trajectory data. Figure 7 As shown, the step of performing denoising on the radar trajectory data to obtain the first trajectory data in the above S200 can be implemented in the following manner:
[0089] S210: Encode the radar trajectory data to obtain radar encoded trajectory data of the target vehicle.
[0090] In practical applications, computer equipment can use a coding algorithm to encode radar trajectory data to obtain radar coded trajectory data of the target vehicle.
[0091] Optionally, the encoding algorithm may be implemented by at least one of a linear extrapolation method, a single-point starting method, a multi-point starting method, a probabilistic data association method, etc. In addition, the encoding algorithm may also be an algorithm constructed based on a deep learning model.
[0092] In an embodiment of the present application, the computer device may input the radar trajectory data into a denoising autoencoder (DAE) network model, and the DAE network model encodes the radar trajectory data and outputs the radar encoded trajectory data of the target vehicle.
[0093] S220: Decode the radar coded trajectory data to obtain first trajectory data.
[0094] Specifically, the computer device may use a decoding algorithm to decode the radar coded trajectory data to obtain the first trajectory data.
[0095] Optionally, the decoding algorithm may be implemented by at least one of a dynamic programming algorithm, a steepest descent method, a heuristic search algorithm, an optimal control method, a randomization algorithm, etc. In addition, the decoding algorithm may also be an algorithm constructed based on a neural network model.
[0096] In an embodiment of the present application, the computer device may use a decoder to decode the radar coded trajectory data to obtain the first trajectory data.
[0097] For example, Figure 8The structure diagram of the denoising processing model is shown, wherein the denoising processing model includes a DAE network model and a decoder connected in series. If x represents radar trajectory data, y represents radar coded trajectory data, and z represents the first trajectory data.
[0098] At the same time, in one embodiment, if Fig. 9 As shown, the step of performing denoising on the AIS trajectory data to obtain the second trajectory data in the above S200 can be implemented in the following manner:
[0099] S230: Encode the AIS trajectory data to obtain AIS encoded trajectory data of the target vehicle.
[0100] In practical applications, the computer device may use a coding algorithm to encode the AIS trajectory data to obtain the AIS coded trajectory data of the target vehicle. The coding algorithm here is similar to the coding algorithm in S210 above, and will not be described in detail.
[0101] In an embodiment of the present application, the computer device may input the AIS trajectory data into the DAE network model, and the DAE network model encodes the AIS trajectory data and then outputs the AIS encoded trajectory data of the target vehicle.
[0102] S240: Decode the AIS encoded trajectory data to obtain second trajectory data.
[0103] Furthermore, the computer device may decode the AIS coded trajectory data using a decoding algorithm to obtain the AIS coded trajectory data of the target vehicle. The decoding algorithm here is similar to the decoding algorithm in S220 above, and will not be described in detail.
[0104] In an embodiment of the present application, the computer device may use a decoder to decode the AIS encoded trajectory data to obtain second trajectory data.
[0105] It should be noted here that the execution order of the above steps S210-S220 and steps S230-S240 can be interchanged, and this execution order is not fixed in the embodiment of the present application.
[0106] The technical solution in the embodiment of the present application can encode and decode the radar trajectory data and the AIS trajectory data in sequence to achieve denoising of the radar trajectory data and the AIS trajectory data, so that the subsequent association detection process has stronger noise resistance. On this basis, subsequent processing can be further performed based on the trajectory data after denoising with higher quality to improve the accuracy of association detection.
[0107] The following is an explanation of the process of determining the trajectory characteristics of the target vehicle by fusing the first trajectory data and the second trajectory data. Fig.10 As shown, the steps in the above S300 can be implemented in the following manner:
[0108] S310 , performing difference processing according to the first trajectory data and the second trajectory data to determine position difference data, speed difference data, and heading difference data between the radar trajectory and the AIS trajectory.
[0109] Specifically, the computer device can pre-train a data fusion model, and then input the first trajectory data and the second trajectory data into the data fusion model. The data fusion model outputs position difference data, speed difference data and heading difference data between the radar trajectory and the AIS trajectory after fusing the first trajectory data and the second trajectory data.
[0110] For example, if the first trajectory data is , the second trajectory data is ,in, Include track location , trajectory speed and trajectory heading , Also includes track position , trajectory speed and trajectory heading .
[0111] Specifically, the computer device can be based on the trajectory position in the first trajectory data and the trajectory position in the second trajectory data Perform difference processing to obtain the position difference data between the radar track and the AIS track , where the position difference data It can be expressed as:
[0112] (3)
[0113] At the same time, the computer device can calculate the trajectory speed according to the trajectory speed in the first trajectory data. and the trajectory speed in the second trajectory data Perform difference processing to obtain the speed difference data between the radar track and the AIS track , where the speed difference data It can be expressed as:
[0114] (4)
[0115] In addition, the computer device can calculate the trajectory heading in the first trajectory data. and the track heading in the second track data Perform difference processing to obtain the heading difference data between the radar track and the AIS track , where the heading difference data It can be expressed as:
[0116] (5)
[0117] It should be noted here that the position difference data, speed difference data and heading difference data are calculated by respectively calculating the track position, track speed and track heading corresponding to the same time in the first track data and the second track data.
[0118] In the embodiment of the present application, the position difference data between the radar track and the AIS track finally obtained is It can be expressed as , speed difference data It can be expressed as , heading difference data It can be expressed as .
[0119] S320: constructing trajectory features of the target vehicle according to the position difference data, the speed difference data and the heading difference data.
[0120] Specifically, the computer device can pre-train an algorithm model, and then input the position difference data, speed difference data and heading difference data between the radar track and the AIS track of the target vehicle into the algorithm model, and the algorithm model outputs the track characteristics of the target vehicle.
[0121] In addition, the computer device can perform conversion processing and / or arithmetic processing on the position difference data, speed difference data and heading difference data between the radar track and the AIS track of the target vehicle to obtain the track characteristics of the target vehicle. Optionally, the arithmetic operation can be implemented by at least one of addition, subtraction, division, logarithm, multiplication, exponential operation, etc.
[0122] In one embodiment, the step in S320 above may include: combining the position difference data, the speed difference data and the heading difference data to generate a trajectory feature.
[0123] In the embodiment of the present application, the computer device can store the position difference data , speed difference data and heading difference data The target vehicle’s trajectory feature D can be expressed as .
[0124] The technical solution in the embodiment of the present application performs difference processing based on the first trajectory data and the second trajectory data, determines the position difference data, speed difference data and heading difference data between the radar trajectory and the AIS trajectory, and constructs the trajectory characteristics of the target vehicle based on the position difference data, speed difference data and heading difference data; the above method can fuse the position difference data, speed difference data and heading difference data between the radar trajectory and the AIS trajectory to form the trajectory characteristics of the target vehicle, so as to provide comprehensive and rich basic data for subsequent association detection and improve the accuracy of association detection.
[0125] In one embodiment, the present application also provides a data association detection method, which is applied to a computer device. The method includes the following process:
[0126] (1) Obtain the original radar trajectory data and original AIS trajectory data of the target vehicle;
[0127] (2) Time alignment processing is performed on the original radar trajectory data and the original AIS trajectory data to obtain radar trajectory data and AIS trajectory data;
[0128] (3) Encoding the radar trajectory data to obtain radar coded trajectory data of the target vehicle;
[0129] (4) decoding the radar coded trajectory data to obtain first trajectory data; and,
[0130] (5) Encoding the AIS trajectory data to obtain the AIS coded trajectory data of the target vehicle;
[0131] (6) Decoding the AIS encoded trajectory data to obtain second trajectory data;
[0132] (7) performing difference processing on the first trajectory data and the second trajectory data to determine position difference data, speed difference data, and heading difference data between the radar trajectory and the AIS trajectory;
[0133] (8) combining the position difference data, the speed difference data, and the heading difference data to generate the trajectory characteristics of the target vehicle;
[0134] (9) The trajectory features are input into a pre-built classification model to perform correlation detection on the radar trajectory and AIS trajectory of the target vehicle, thereby obtaining the correlation status between the radar trajectory and AIS trajectory of the target vehicle.
[0135] The execution process of the above (1) to (9) can be specifically referred to the description of the above embodiment. The implementation principle and technical effect are similar and will not be repeated here.
[0136] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0137] Based on the same inventive concept, the embodiment of the present application also provides a data association detection device for implementing the data association detection method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more data association detection device embodiments provided below can refer to the limitations on the data association detection method above, and will not be repeated here.
[0138] In one embodiment, Fig.11 This is a schematic diagram of the structure of a data association detection device in one embodiment of the present application. The data association detection device provided in the embodiment of the present application can be applied to a computer device. Fig.11 As shown, the data association detection device of the embodiment of the present application may include: an acquisition module 11, a denoising processing module 12, a determination module 13 and an association detection module 14, wherein:
[0139] An acquisition module 11 is used to acquire radar track data and ship automatic identification system AIS track data of the target vehicle;
[0140] A denoising processing module 12 is used to perform denoising processing on the radar trajectory data to obtain first trajectory data; and to perform denoising processing on the AIS trajectory data to obtain second trajectory data;
[0141] A determination module 13, configured to perform a fusion process based on the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle;
[0142] The association detection module 14 is used to input the trajectory features into a pre-built classification model to perform association detection on the radar trajectory and the AIS trajectory of the target vehicle, and obtain the association status between the radar trajectory and the AIS trajectory of the target vehicle.
[0143] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0144] In one embodiment, the acquisition module 11 is specifically used for:
[0145] Obtain the original radar track data and original AIS track data of the target vehicle;
[0146] The original radar track data and the original AIS track data are time aligned to obtain radar track data and AIS track data.
[0147] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0148] In one embodiment, the denoising processing module 12 includes: a first encoding processing unit and a first decoding processing unit, wherein:
[0149] A first encoding processing unit, used for encoding the radar trajectory data to obtain radar encoding trajectory data of the target vehicle;
[0150] The first decoding processing unit is used to decode the radar coded trajectory data to obtain first trajectory data.
[0151] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0152] In one embodiment, the denoising processing module 12 includes: a second encoding processing unit and a second decoding processing unit, wherein:
[0153] A second encoding processing unit is used to encode the AIS trajectory data to obtain the AIS encoded trajectory data of the target vehicle;
[0154] The second decoding processing unit is used to decode the AIS encoded trajectory data to obtain second trajectory data.
[0155] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0156] In one embodiment, the determination module 13 includes: a determination unit and a construction unit, wherein:
[0157] a determining unit, configured to perform difference processing according to the first track data and the second track data, and determine position difference data, speed difference data and heading difference data between the radar track and the AIS track;
[0158] The construction unit is used to construct the trajectory characteristics of the target vehicle according to the position difference data, the speed difference data and the heading difference data.
[0159] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0160] In one embodiment, the building block is specifically used for:
[0161] The position difference data, speed difference data and heading difference data are combined to generate trajectory features.
[0162] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0163] In one embodiment, the association detection module 14 is specifically used for:
[0164] Obtain trajectory feature sets and trajectory data sets of different vehicles; the trajectory data sets include multiple labeled radar trajectory data and labeled AIS trajectory data;
[0165] According to the trajectory data set and the trajectory data set, the initial classification model is trained to obtain a trained classification model.
[0166] The data association detection device provided in the embodiment of the present application can be used to execute the technical solution in the above-mentioned data association detection method embodiment of the present application. Its implementation principle and technical effect are similar and will not be repeated here.
[0167] For the specific definition of the data association detection device, please refer to the definition of the data association detection method above, which will not be repeated here. Each module in the above-mentioned data association detection device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0168] In one embodiment, a computer device is also provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.12As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide processing capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store radar trajectory data and AIS trajectory data. The network interface of the computer device is used to communicate with an external end point through a network connection. When the computer program is executed by the processor, a data association detection method is implemented.
[0169] Those skilled in the art will understand that Fig.12 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0170] In one embodiment, a computer device is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the technical solution in the above-mentioned data association detection method embodiment of the present application is implemented. The implementation principle and technical effect are similar and will not be repeated here.
[0171] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the technical solution of the above-mentioned data association detection method of the present application is implemented. The implementation principle and technical effect are similar and will not be repeated here.
[0172] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the technical solution of the above-mentioned data association detection method of the present application is implemented. The implementation principle and technical effect are similar and will not be repeated here.
[0173] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0174] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0175] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A data association detection method, characterized in that: The method comprises: Obtain radar track data and ship automatic identification system AIS track data of target vehicles; Performing denoising processing on the radar trajectory data to obtain first trajectory data; and performing denoising processing on the AIS trajectory data to obtain second trajectory data; Performing fusion processing on the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle; The trajectory features are input into a pre-built classification model to perform correlation detection on the radar trajectory and the AIS trajectory of the target vehicle, so as to obtain a correlation state between the radar trajectory and the AIS trajectory of the target vehicle.
2. The method according to claim 1, characterized in that The step of obtaining radar track data and automatic ship identification system AIS track data of the target vehicle includes: Acquiring original radar track data and original AIS track data of the target vehicle; The original radar track data and the original AIS track data are time aligned to obtain the radar track data and the AIS track data.
3. The method according to claim 1 or 2, characterized in that: The performing denoising processing on the radar trajectory data to obtain first trajectory data includes: Encoding the radar trajectory data to obtain the radar coded trajectory data of the target vehicle; The radar encoded trajectory data is decoded to obtain the first trajectory data.
4. The method according to claim 1 or 2, characterized in that: The performing denoising processing on the AIS trajectory data to obtain second trajectory data includes: Encoding the AIS trajectory data to obtain the AIS coded trajectory data of the target vehicle; The AIS encoded trajectory data is decoded to obtain the second trajectory data.
5. The method according to claim 1 or 2, characterized in that: The step of performing fusion processing on the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle includes: Performing difference processing on the first trajectory data and the second trajectory data to determine position difference data, speed difference data and heading difference data between the radar trajectory and the AIS trajectory; The trajectory characteristics of the target vehicle are constructed according to the position difference data, the speed difference data and the heading difference data.
6. The method according to claim 5, characterized in that The step of constructing the trajectory feature of the target vehicle according to the position difference data, the speed difference data and the heading difference data comprises: The position difference data, the speed difference data and the heading difference data are combined to generate the trajectory feature.
7. The method according to claim 1 or 2, characterized in that: The construction process of the classification model includes: Acquire trajectory feature sets and trajectory data sets of different vehicles; the trajectory data sets include a plurality of labeled radar trajectory data and labeled AIS trajectory data; According to the trajectory data set and the trajectory data set, an initial classification model is trained to obtain a trained classification model.
8. A data association detection device, characterized in that: The device comprises: An acquisition module is used to acquire radar track data of target vehicles and ship automatic identification system AIS track data; A denoising processing module, configured to perform denoising processing on the radar trajectory data to obtain first trajectory data; and perform denoising processing on the AIS trajectory data to obtain second trajectory data; a determination module, configured to perform fusion processing on the first trajectory data and the second trajectory data to determine the trajectory characteristics of the target vehicle; The association detection module is used to input the trajectory feature into a pre-built classification model to perform association detection on the radar trajectory and the AIS trajectory of the target vehicle, and obtain the association state between the radar trajectory and the AIS trajectory of the target vehicle.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.