A track segment-based preprocessing method and device
By segmenting and preprocessing the track, calculating key points and extracting 11-dimensional feature information, the problem of large data accumulation in long-term tracking scenarios is solved, and track data compression and feature extraction are realized, which is suitable for track data analysis and target recognition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NAVAL AVIATION UNIV
- Filing Date
- 2024-12-25
- Publication Date
- 2026-06-16
Smart Images
Figure CN119884113B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a preprocessing method and apparatus based on track segmentation, belonging to the field of track data processing technology. Background Technology
[0002] Track data is acquired frame-by-frame based on sensor sampling frequency, resulting in massive data accumulation in long-term tracking scenarios. In real-world applications, some tasks do not require all track points to participate in calculations, but rather focus on feature information over a long time span. Therefore, preprocessing of the accumulated raw track data is necessary to form a standardized feature matrix. This mainly involves track segmentation techniques and feature extraction techniques.
[0003] The navigation of a target typically comprises multiple phases, each exhibiting distinct characteristics, with significant differences between navigation within operational and non-operational areas. Therefore, track segmentation is necessary, preserving the fundamental features of each segment. This segmentation requires retaining key points that maintain the overall track profile. A typical track segmentation technique is the Douglas-Peucker (DP) algorithm; however, in real-world applications, it's crucial to balance sufficient track compression with minimal retention of maneuver details. The threshold settings in the classic DP algorithm struggle to reconcile these two requirements. Furthermore, track preprocessing research lacks a process for extracting fundamental feature information from individual track segments. Summary of the Invention
[0004] To address the problems existing in the prior art, this invention provides a preprocessing method and apparatus based on track segmentation. The track preprocessing method and apparatus of this invention compress the track and obtain a normalized feature tensor by performing segmentation preprocessing on the track. The specific technical solution is as follows:
[0005] A preprocessing method based on track segmentation includes:
[0006] Step 1: Read the track data for a single target batch;
[0007] Step 2: Calculate the keypoint sets in the longitude and latitude directions respectively. Each keypoint contains three-dimensional information about its track, namely its index value, longitude value, and latitude value. Take the union of the two sets to obtain the track keypoint set.
[0008] Step 3: Divide the entire track into segments based on the set of key points to obtain a set of segmented tracks;
[0009] Step 4: Process each track segment separately, calculate the feature information of each track segment, obtain the 11-dimensional information of each track segment, arrange the retained information of each track segment in chronological order, and obtain the information tensor.
[0010] Furthermore, step 2 specifically includes:
[0011] Step 2.1: Normalize the longitude, latitude, and time of the flight path respectively;
[0012] Step 2.2: This step performs calculations in the normalized longitude-time two-dimensional space, using Euclidean distance as the metric. The first and last points of the track are added to the "longitude-time" track key point set. Other points on the track are traversed to obtain the point in this space that is farthest from the line connecting the first and last endpoints of the track, along with its farthest distance. This point and its index are added to the "longitude-time" track key point set. The key points in the set divide the track into two segments. The points in each segment are traversed, and the point that is farthest from the line connecting the first and last endpoints of the segment, along with its farthest distance, is calculated. This point is added to the "longitude-time" track key point set. This process is repeated until all farthest distances are less than a set threshold, resulting in the final "longitude-time" track key point set.
[0013] Step 2.3: This step is performed in the normalized latitude-time two-dimensional space, using Euclidean distance as the metric. The operations performed in this space are the same as those in the longitude-time two-dimensional space described in Step 2.2, and a set of key points for the latitude-time track is obtained.
[0014] Step 2.4: Take the union of the "longitude-time" track key point set and the "latitude-time" track key point set to obtain the final track key point set.
[0015] Furthermore, step 4 specifically includes:
[0016] Step 4.1: Read each track segment from Step 3, and calculate the duration, length of the endpoint connection line, azimuth angle of the endpoint connection line, cumulative number of points, average speed, longitude difference value, and latitude difference value of each track segment. In addition, retain the latitude and longitude of the endpoints.
[0017] Step 4.2: Arrange the 11-dimensional information of each obtained track segment to obtain the information tensor of the original track.
[0018] Furthermore, in step 4.1, the specific process for calculating the information in each dimension is as follows:
[0019] Step 4.1.1: Calculate the duration of each track segment based on the time values at the endpoints of each track segment;
[0020] Step 4.1.2: Calculate the length of the line connecting the endpoints of each track segment based on the latitude and longitude values of each track segment endpoint;
[0021] Step 4.1.3: Calculate the azimuth of the line connecting the endpoints of each track segment based on the latitude and longitude values of each track segment endpoint;
[0022] Step 4.1.4: Calculate the number of original points contained in each track segment to obtain the cumulative number of points;
[0023] Step 4.1.5: Calculate the average speed of each track segment based on the latitude and longitude values of all track points in each track segment;
[0024] Step 4.1.6: Calculate the longitude difference value and latitude difference value based on the latitude and longitude values of the endpoints of each flight path segment.
[0025] The present invention provides a preprocessing apparatus based on track segmentation, characterized in that it includes:
[0026] The trajectory information recording module allows you to input the trajectory data of the target in the current batch and choose to end trajectory information recording during the tracking process or when the batch ends, depending on the mission requirements.
[0027] The track key point calculation module includes reading data from the track information acquisition module, obtaining the "longitude-time" track key point set and the "latitude-time" track key point set, and taking the union of them to obtain the track key point set;
[0028] The track segmentation module completes the segmented storage of the track, divides the entire track into segments based on the track's key points, and establishes an index for each segment.
[0029] The track information tensor calculation module completes the calculation of the track information tensor. It reads each segment of track data separately, calculates 11-dimensional information for each segment, and finally merges them to obtain the track information tensor.
[0030] The trajectory preprocessing method and apparatus of this invention compresses and normalizes the trajectory by performing segmented preprocessing, thereby obtaining a normalized feature tensor. Compared with the prior art, the beneficial effects of this invention are: it can solve the data storage burden problem caused by the large amount of data accumulation in long-term tracking scenarios; it can reduce data redundancy by obtaining normalized feature vectors through segmented preprocessing; it can effectively extract key points of the trajectory and segment the trajectory based on these points, and the key points can not only be used for segmentation but also basically preserve the overall picture of the trajectory; the feature tensor obtained by extracting feature information from each trajectory segment can be used in tasks such as trajectory data analysis and target recognition, and has high engineering application value. Attached Figure Description
[0031] Figure 1 This is a schematic diagram of a preprocessing method based on track segmentation according to the present invention;
[0032] Figure 2 This is a schematic diagram of the preprocessing device based on track segmentation according to the present invention. Detailed Implementation
[0033] To more clearly illustrate the technical solution of the present invention, the technical solution of the present invention will be described in detail below with reference to the embodiments and accompanying drawings.
[0034] To preprocess the accumulated raw track data and form a normalized feature matrix, this invention provides a track preprocessing method, such as... Figure 1 As shown, the method includes the following steps:
[0035] Step 1: Read the track data for a single target batch: T = {P1, ..., Pj, ..., Pm}, j ∈ [1, m]; where T represents the current track, P... j This represents the j-th point in the trajectory.
[0036] Step 2: Calculate the keypoint sets in the longitude and latitude directions respectively. Each keypoint contains three-dimensional information about its track, namely its index value, longitude value, and latitude value. Take the union of the two sets to obtain the track keypoint set.
[0037] Step 2.1: Normalize the longitude, latitude, and time of the flight track respectively. The normalization method includes, but is not limited to, the normalization method used in this embodiment:
[0038]
[0039]
[0040]
[0041] In the formula, lon j lat j t j Divided into the longitude, latitude, and time of the j-th track point, lon max lon min lat max lat min t max t min These represent the maximum and minimum longitude, maximum and minimum latitude, maximum and minimum time values for all track points in track T.
[0042] Step 2.2: This step performs calculations in the normalized longitude-time two-dimensional space, using Euclidean distance as the metric. The first and last points of the track are added to the "longitude-time" track key point set. Other points on the track are traversed to obtain the point in this space that is farthest from the line connecting the first and last endpoints of the track, along with its farthest distance. This point and its index are added to the "longitude-time" track key point set. The key points in the set divide the track into two segments. The points in each segment are traversed, and the point that is farthest from the line connecting the first and last endpoints of the segment, along with its farthest distance, is calculated. This point is added to the "longitude-time" track key point set. This process is repeated until all farthest distances are less than a set threshold, resulting in the final "longitude-time" track key point set.
[0043] Step 2.2.1: First, set the first and last two track points P1 and P2 of the track. m Add to the "longitude-time" track key point set A, traverse other points in the track, and calculate the point P for each point in the longitude-time two-dimensional space. j j∈[2,m-1] to line segment P1P m European distance And obtain the maximum distance.
[0044]
[0045] In the formula,
[0046]
[0047] And obtain the maximum distance:
[0048]
[0049] Step 2.2.2: Determine if the maximum distance is greater than the set threshold d. thre If the distance is greater than the given distance, the track point at the maximum distance is added to set A; otherwise, the key point search ends.
[0050] Step 2.2.3: After steps 2.2.1 and 2.2.2 are completed, the track key points in set A divide the track into several track segments. Each track segment is treated as a track, and steps 2.2.1 and 2.2.2 are repeated until the maximum distance d of each segment is reached. max Not exceeding the threshold d thre This yields the final set of key points A for the "longitude-time" trajectory.
[0051] Step 2.3: This step is performed in the normalized latitude-time two-dimensional space, using Euclidean distance as the metric. The operations performed in this space are the same as those in the longitude-time two-dimensional space described in Step 2.2, and a set of key points for the latitude-time track is obtained.
[0052] Step 2.3.1: First, mark the first and last two track points P1 and P2 of the track. m Add to the "longitude-time" track key point set B, traverse other points in the track, and calculate the point P for each point in the latitude-time two-dimensional space. j j∈[2,m-1] to line segment P1P m European distance And obtain the maximum distance.
[0053]
[0054] In the formula,
[0055]
[0056] And obtain the maximum distance:
[0057]
[0058] Step 2.3.2: Determine if the maximum distance is greater than the set threshold d. thre If the distance is greater than the given distance, the track point at the maximum distance is added to set B; otherwise, the key point search ends.
[0059] Step 2.3.3: After steps 2.2.1 and 2.2.2 are completed, the track key points in set B divide the track into several track segments. Each track segment is treated as a track, and steps 2.2.1 and 2.2.2 are repeated until the maximum distance d of each segment is reached. max Not exceeding the threshold d thre This yields the final set of key points B for the "latitude-time" track.
[0060] Step 2.4: Take the union of the "longitude-time" track key point set and the "latitude-time" track key point set to obtain the final track key point set C.
[0061] Step 3: Divide the entire track into segments based on the set of key points to obtain a set of segmented tracks.
[0062] TR = {TR1, TR2, TR3, ..., TR} n} (10)
[0063] Step 4: Process each track segment separately, calculate the feature information of each track segment, obtain the 11-dimensional information of each track segment, arrange the retained information of each track segment in chronological order, and obtain the information tensor.
[0064] Step 4.1: Read each track segment from Step 3, and calculate the duration, length of the endpoint connection line, azimuth angle of the endpoint connection line, cumulative number of points, average speed, longitude difference value, and latitude difference value of each track segment. In addition, retain the latitude and longitude of the endpoints.
[0065] Step 4.1.1: Calculate the duration of each track segment based on the time values at the endpoints of each track segment;
[0066] λ i1 =t imax -t imin (11)
[0067] Step 4.1.2: Calculate the length of the line connecting the endpoints of each track segment based on the latitude and longitude values of each track segment endpoint;
[0068]
[0069] Step 4.1.3: Calculate the azimuth of the line connecting the endpoints of each track segment based on the latitude and longitude values of each track segment endpoint;
[0070]
[0071] Step 4.1.4: Calculate the number of original points contained in each track segment to obtain the cumulative number of points;
[0072] λ i4 =card(TR) i (14)
[0073] Step 4.1.5: Calculate the average speed of each track segment based on the latitude and longitude values of all track points in each track segment;
[0074]
[0075] Step 4.1.6: Calculate the longitude difference value and latitude difference value based on the latitude and longitude values of the endpoints of each flight path segment.
[0076] λ i6 =lon imax -lon imin (16)
[0077] λ i7 =lat imax -lat imin (17)
[0078] Step 4.1.7: Retain the latitude and longitude of the endpoints.
[0079] λ i8 =lon imax (18)
[0080] λi9 =lat imax (19)
[0081] λ i10 =lon imin (20)
[0082] λ i11 =lat imin (twenty one)
[0083] Step 4.2: Arrange the 11-dimensional information of each track segment obtained in Step 4.2 to obtain the feature tensor of the original track.
[0084]
[0085] like Figure 2 As shown, a preprocessing device based on track segmentation includes:
[0086] The trajectory information recording module allows you to input the trajectory data of the target in the current batch and choose to end trajectory information recording during the tracking process or when the batch ends, depending on the mission requirements.
[0087] The track key point calculation module includes reading data from the track information acquisition module, obtaining the "longitude-time" track key point set and the "latitude-time" track key point set, and taking the union of them to obtain the track key point set;
[0088] The track segmentation module completes the segmented storage of the track, divides the entire track into segments based on the track's key points, and establishes an index for each segment.
[0089] The track information tensor calculation module completes the calculation of the track information tensor. It reads each segment of track data separately, calculates 11-dimensional information for each segment, and finally merges them to obtain the track information tensor.
[0090] The trajectory preprocessing method and apparatus of the present invention preprocesses the trajectory in segments, compresses the trajectory, and obtains a normalized feature tensor.
Claims
1. A preprocessing method based on track segmentation, characterized in that... include: Step 1: Read the track data for a single target batch; Step 2: Calculate the key point sets in the longitude and latitude directions respectively. The key points contain the three-dimensional information of the point tracks, which are the index value, longitude value and latitude value respectively. Take the union of the two sets to obtain the track key point set. Step 3: Divide the entire track into segments based on the set of key points to obtain a set of segmented tracks; Step 4: Process each track segment separately, calculate the feature information of each track segment, obtain the 11-dimensional information of each track segment, arrange the retained information of each track segment in chronological order, and obtain the information tensor; Step 2 specifically includes: Step 2.1: Normalize the longitude, latitude, and time of the flight path respectively; Step 2.2: This step performs calculations in the normalized longitude-time two-dimensional space, using Euclidean distance as the metric. The first and last points of the track are added to the "longitude-time" track key point set. Other points on the track are traversed to obtain the point in this space that is farthest from the line connecting the first and last endpoints of the track, along with its farthest distance. This point and its index are added to the "longitude-time" track key point set. The key points in the set divide the track into two segments. The points in each segment are traversed, and the point that is farthest from the line connecting the first and last endpoints of the segment, along with its farthest distance, is calculated. This point is added to the "longitude-time" track key point set. This process is repeated until all farthest distances are less than a set threshold, resulting in the final "longitude-time" track key point set. Step 2.3: This step is performed in the normalized latitude-time two-dimensional space, using Euclidean distance as the metric. The operations performed in this space are the same as those in the longitude-time two-dimensional space in Step 2.2, and a set of key points for the "latitude-time" track is obtained. Step 2.4: Take the union of the "longitude-time" track key point set and the "latitude-time" track key point set to obtain the final track key point set; Step 4 specifically includes: Step 4.1: Read each track segment from Step 3, and calculate the duration, length of the endpoint connection line, azimuth angle of the endpoint connection line, cumulative number of points, average speed, longitude difference value, and latitude difference value of each track segment. In addition, retain the latitude and longitude of the endpoints. Step 4.2: Arrange the 11-dimensional information of each obtained track segment to obtain the information tensor of the original track; In step 4.1, the specific process for calculating the information of each dimension is as follows: Step 4.1.1: Calculate the duration of each track segment based on the time values at the endpoints of each track segment; Step 4.1.2: Calculate the length of the line connecting the endpoints of each track segment based on the latitude and longitude values of each track segment endpoint; Step 4.1.3: Calculate the azimuth of the line connecting the endpoints of each track segment based on the latitude and longitude values of each track segment endpoint; Step 4.1.4: Calculate the number of original points contained in each track segment to obtain the cumulative number of points; Step 4.1.5: Calculate the average speed of each track segment based on the latitude and longitude values of all track points in each track segment; Step 4.1.6: Calculate the longitude difference value and latitude difference value based on the latitude and longitude values of the endpoints of each flight path segment.
2. An apparatus for implementing the preprocessing method based on track segmentation as described in claim 1, characterized in that... include: The trajectory information recording module allows you to input the trajectory data of the target in the current batch and choose to end trajectory information recording during the tracking process or when the batch ends, depending on the mission requirements. The track key point calculation module includes reading data from the track information acquisition module, obtaining the "longitude-time" track key point set and the "latitude-time" track key point set, and taking the union of them to obtain the track key point set; The track segmentation module completes the segmented storage of the track, divides the entire track into segments based on the track's key points, and establishes an index for each segment; The track information tensor calculation module completes the calculation of the track information tensor. It reads each segment of track data separately, calculates 11-dimensional information for each segment, and finally merges them to obtain the track information tensor.