CN113011520A - Radar data filtering method based on thermodynamic diagram and graph theory clustering - Google Patents
Radar data filtering method based on thermodynamic diagram and graph theory clustering Download PDFInfo
- Publication number
- CN113011520A CN113011520A CN202110355748.3A CN202110355748A CN113011520A CN 113011520 A CN113011520 A CN 113011520A CN 202110355748 A CN202110355748 A CN 202110355748A CN 113011520 A CN113011520 A CN 113011520A
- Authority
- CN
- China
- Prior art keywords
- radar data
- points
- area
- radar
- thermodynamic diagram
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2323—Non-hierarchical techniques based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/904—Browsing; Visualisation therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Discrete Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a radar data filtering method based on thermodynamic diagram and graph theory clustering, which comprises the following steps: the method comprises the following steps: radar data are obtained and cached, and the radar data are mapped into a Heat Map to generate a corresponding thermodynamic diagram; step two: analyzing the frequency and the condition of the point in each area by observing the thermodynamic diagram, setting a threshold value, and finding and marking all the points near the threshold value; step three: after a boundary threshold point set needing to be processed is obtained, finding out the boundaries of the points through Alpha Shapes to obtain a fixed noise point generating area; step four: after the Alpha Shapes algorithm calculates and obtains boundary point cluster sets, whether points thrown out by the radar fall in the boundary point cluster sets or not is judged, and radar data preprocessing is completed. The method is suitable for processing the vertex radar data of multiple noise points in the complex environment, has good adaptability to the mobile radar data or the radar data in the sudden change environment by automatically identifying the noise source area in real time, and is high in identification efficiency, simple and effective.
Description
Technical Field
The invention relates to the field of data preprocessing, in particular to a radar data filtering method based on thermodynamic diagram and graph theory clustering.
Background
In recent years, the technology of the civil unmanned aerial vehicle is mature day by day, the civil unmanned aerial vehicle is applied more and more widely, and meanwhile, the control accident and related safety problems caused by the civil unmanned aerial vehicle are normalized day by day, and the civil unmanned aerial vehicle can be used more and more maliciously in a plurality of beneficial application scenes.
This is the case for the slow and slow drone counter-braking system. The low-small-slow unmanned aerial vehicle countercheck system mainly comprises a detection part and a countercheck part, and a radar in the detection part plays an important detection role, so that the target screening of radar data is greatly challenged. Too many noise points in a fixed area will affect the detection of real targets by radar, and it is necessary to filter the fixed noise sources in a complex environment.
For example, a method for labeling a sample of radar signal detection data disclosed in the published patent (application number CN 202011048313.6) mainly aims at the problem that the existing sample labeling method is a method for analyzing and extracting radar data one by one through manual labeling, and the method has high requirements on the quality of labeling personnel, large workload and low working efficiency. The method comprises the following steps: extracting effective data through visual software for marking, and storing the effective data as a standard sample file; and traversing and acquiring the original data file by taking the standard file as a template, and automatically generating a new sample file. The extraction of the radar signal detection data of this patent needs to consume a large amount of manpowers, and data calculation time consuming is longer, so is not efficient in practical application.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a radar data filtering method based on thermodynamic diagram and graph theory clustering.
The purpose of the invention is realized by the following technical scheme:
the radar data filtering method based on thermodynamic diagram and graph theory clustering comprises the following steps:
the method comprises the following steps: radar data are obtained and cached, and the radar data are mapped into a Heat Map to generate a corresponding thermodynamic diagram;
step two: analyzing the frequency and the condition of the point in each area by observing the thermodynamic diagram, setting a threshold value, and finding and marking all the points near the threshold value;
step three: after a boundary threshold point set needing to be processed is obtained, finding out the boundaries of the points through Alpha Shapes to obtain a fixed noise point generating area;
step four: after the Alpha Shapes algorithm calculates and obtains boundary point cluster sets, whether points thrown out by the radar fall in the boundary point cluster sets or not is judged, and radar data preprocessing is completed.
The first step specifically comprises the following steps:
assuming that all radar data obey Gaussian distribution, setting a region with the size of 1600 x 1600, and setting the width and standard deviation sigma of a Gaussian kernel, wherein the ratio of the width to the standard deviation sigma of the Gaussian kernel is 3;
mapping the longitude and latitude of the radar to the center of the area, and mapping the radar data to the area with a scale of 1: 100, namely mapping a range area of 8km near the radar into the area, and giving the same weight to the mapped radar data points;
radar points are placed in a set mapping area, a weighted value of a nearby area is calculated through a Gaussian kernel function when each point is placed, and the weight of each point in the area can be obtained after all radar data are placed in the block area;
traversing the whole mapping area to obtain the value with the maximum area weight, setting a gradual change color bar, wherein the corresponding weight is from 0 to the maximum weight, and then performing color rendering of the corresponding weight on each point of the mapping area to obtain the thermodynamic diagram distribution of the radar data.
The second threshold calibration method specifically comprises the following steps:
assuming that 3 targets appear in the range of 50m and are considered as a noise source area, 3 radar data with a distance of 50m are mapped into a mapping area, and whether a weight capable of containing all the 3 points exists is observed through a generated visual thermodynamic diagram, and if the weight exists, the weight is used as a noise source area threshold value.
The Alpha Shapes algorithm flow specifically comprises the following steps:
setting a judgment radius R, and assuming that n disordered points P1 and P2 … Pn are arranged near a marking threshold;
drawing a circle with the radius R through any two points P1 and P2, if no other data points exist in the circle, considering the points P1 and P2 as boundary points, and a connecting line P1P2 of the points is a boundary line segment;
the n data points are connected pairwise to form (n x (n-1))/2 line segments, and judgment and solution are carried out one by one.
The judging method of the fourth step specifically comprises the following steps: judging whether the points thrown out by the radar fall in the boundary point cluster sets or not for each boundary cluster; if the radar data points are in the set, marking the radar data points as environmental noise points; if the radar data point is not in the set, the radar data point is marked as a valid data point.
The cached radar data is provided with a valid period, and the data is invalidated once the valid period is exceeded.
The invention has the beneficial effects that:
the method and the device automatically identify the noise source area in real time, so the method and the device have good adaptability to the mobile radar data or the radar data in a sudden change environment, have high identification efficiency, and are simple and effective.
Drawings
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a thermodynamic diagram after screening in the present invention;
FIG. 3 is a threshold point set boundary diagram of the present invention.
Detailed Description
In order to more clearly understand the technical features, objects, and effects of the present invention, embodiments of the present invention will now be described with reference to the accompanying drawings.
In this embodiment, as shown in fig. 1, the radar data filtering method based on thermodynamic diagram and graph theory clustering includes the following steps:
the method comprises the following steps: radar data are obtained and cached, and the radar data are mapped into a Heat Map to generate a corresponding thermodynamic diagram;
step two: analyzing the frequency and the condition of the points in each area by observing the thermodynamic diagram, setting a threshold, finding and marking all the points near the threshold, and obviously seeing the circled positions by visual observation after being taken out as shown in FIG. 2;
step three: after a boundary threshold point set needing to be processed is obtained, finding out the boundaries of the points through Alpha Shapes to obtain a fixed noise point generating area;
step four: after the Alpha Shapes algorithm calculates and obtains boundary point cluster sets, whether points thrown out by the radar fall in the boundary point cluster sets or not is judged, and radar data preprocessing is completed.
The first step specifically comprises the following steps:
assuming that all radar data obey Gaussian distribution, setting a region with the size of 1600 x 1600, and setting the width and standard deviation sigma of a Gaussian kernel, wherein the ratio of the width to the standard deviation sigma of the Gaussian kernel is 3;
mapping the longitude and latitude of the radar to the center of the area, and mapping the radar data to the area with a scale of 1: 100, namely mapping a range area of 8km near the radar into the area, and giving the same weight to the mapped radar data points;
radar points are placed in a set mapping area, a weighted value of a nearby area is calculated through a Gaussian kernel function when each point is placed, and the weight of each point in the area can be obtained after all radar data are placed in the block area;
traversing the whole mapping area to obtain the value with the maximum area weight, setting a gradual change color bar, wherein the corresponding weight is from 0 to the maximum weight, and then performing color rendering of the corresponding weight on each point of the mapping area to obtain the thermodynamic diagram distribution of the radar data.
The second threshold calibration method specifically comprises the following steps:
assuming that 3 targets appear in the range of 50m and are considered as a noise source area, 3 radar data with a distance of 50m are mapped into a mapping area, and whether a weight capable of containing all the 3 points exists is observed through a generated visual thermodynamic diagram, and if the weight exists, the weight is used as a noise source area threshold value.
The threshold value of the proportion can be fixed according to the weight proportion division of the radar data. For example, 5% of the maximum weight value in the mapping region (where the radar data has been mapped) is used as the noise region threshold.
In the second step, a certain range is set for the threshold, and if the threshold is 0.05, the range of the threshold is assumed to be 0.05 ± 0.01, and all the points in the mapping area are traversed to obtain all the points with the weights within the range of the threshold. These points are the desired noise source boundary points. And inversely mapping the noise source into longitude and latitude, and then, obtaining the longitude and latitude boundary of the corresponding actual noise source.
The Alpha Shapes algorithm flow specifically comprises the following steps:
setting a judgment radius R, and assuming that n disordered points P1 and P2 … Pn are arranged near a marking threshold;
drawing a circle with the radius R through any two points P1 and P2, if no other data points exist in the circle, considering the points P1 and P2 as boundary points, and a connecting line P1P2 of the points is a boundary line segment;
the n data points are connected pairwise to form (n x (n-1))/2 line segments, and judgment and solution are carried out one by one.
For obtaining the boundary points, the method adopts Alpha Shapes to realize the extraction of the contour, and the edge of the boundary points can be further extracted by controlling the threshold value of the algorithm. Alpha Shapes represent the boundaries of a set of unordered spatial points, which are not necessarily convex and are not necessarily connected, but which represent, to some extent, the contours of the set of discrete points. The boundary can be made finer or coarser by adjusting the parameters. As shown in fig. 3, (a) is the scatter of the required boundary, and (b) is the boundary obtained by the convex hull algorithm, it is obvious that the graph (c) is the most ideal edge effect.
The judging method of the fourth step specifically comprises the following steps: judging whether the points thrown out by the radar fall in the boundary point cluster sets or not for each boundary cluster; if the radar data points are in the set, marking the radar data points as environmental noise points; if the radar data point is not in the set, the radar data point is marked as a valid data point.
The cached radar data is provided with an effective period, and the data is invalidated once exceeding the effective period, so that the filtering effect of the radar data is prevented from being influenced by excessive environmental factors (cloud and rain weather and the like), and the real-time performance is ensured.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (6)
1. The radar data filtering method based on thermodynamic diagram and graph theory clustering is characterized by comprising the following steps of:
the method comprises the following steps: radar data are obtained and cached, and the radar data are mapped into a Heat Map to generate a corresponding thermodynamic diagram;
step two: analyzing the frequency and the condition of the point in each area by observing the thermodynamic diagram, setting a threshold value, and finding and marking all the points near the threshold value;
step three: after a boundary threshold point set needing to be processed is obtained, finding out the boundaries of the points through Alpha Shapes to obtain a fixed noise point generating area;
step four: after the Alpha Shapes algorithm calculates and obtains boundary point cluster sets, whether points thrown out by the radar fall in the boundary point cluster sets or not is judged, and radar data preprocessing is completed.
2. The radar data filtering method based on thermodynamic diagram and graph theory clustering as claimed in claim 1, wherein the step one is as follows:
assuming that all radar data obey Gaussian distribution, setting a region with the size of 1600 x 1600, and setting the width and standard deviation sigma of a Gaussian kernel, wherein the ratio of the width to the standard deviation sigma of the Gaussian kernel is 3;
mapping the longitude and latitude of the radar to the center of the area, and mapping the radar data to the area with a scale of 1: 100, namely mapping a range area of 8km near the radar into the area, and giving the same weight to the mapped radar data points;
radar points are placed in a set mapping area, a weighted value of a nearby area is calculated through a Gaussian kernel function when each point is placed, and the weight of each point in the area can be obtained after all radar data are placed in the block area;
traversing the whole mapping area to obtain the value with the maximum area weight, setting a gradual change color bar, wherein the corresponding weight is from 0 to the maximum weight, and then performing color rendering of the corresponding weight on each point of the mapping area to obtain the thermodynamic diagram distribution of the radar data.
3. The radar data filtering method based on thermodynamic diagram and graph theory clustering as claimed in claim 1, wherein the step two-threshold calibration method specifically comprises:
assuming that 3 targets appear in the range of 50m and are considered as a noise source area, 3 radar data with a distance of 50m are mapped into a mapping area, and whether a weight capable of containing all the 3 points exists is observed through a generated visual thermodynamic diagram, and if the weight exists, the weight is used as a noise source area threshold value.
4. The radar data filtering method based on thermodynamic diagram and graph theory clustering according to claim 1, wherein the Alpha Shapes algorithm flow specifically is as follows:
setting a judgment radius R, and assuming that n disordered points P1 and P2 … Pn are arranged near a marking threshold;
drawing a circle with the radius R through any two points P1 and P2, if no other data points exist in the circle, considering the points P1 and P2 as boundary points, and a connecting line P1P2 of the points is a boundary line segment;
the n data points are connected pairwise to form (n x (n-1))/2 line segments, and judgment and solution are carried out one by one.
5. The radar data filtering method based on thermodynamic diagram and graph theory clustering as claimed in claim 1, wherein the judging method of the fourth step specifically comprises: judging whether the points thrown out by the radar fall in the boundary point cluster sets or not for each boundary cluster; if the radar data points are in the set, marking the radar data points as environmental noise points; if the radar data point is not in the set, the radar data point is marked as a valid data point.
6. The method for radar data filtering based on thermodynamic diagram and graph theory clustering according to claim 1, wherein the cached radar data is provided with a valid period, and the data is invalidated once the valid period is exceeded.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110355748.3A CN113011520B (en) | 2021-04-01 | 2021-04-01 | Radar data filtering method based on thermodynamic diagram and graph theory clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110355748.3A CN113011520B (en) | 2021-04-01 | 2021-04-01 | Radar data filtering method based on thermodynamic diagram and graph theory clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113011520A true CN113011520A (en) | 2021-06-22 |
| CN113011520B CN113011520B (en) | 2022-11-25 |
Family
ID=76387740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110355748.3A Active CN113011520B (en) | 2021-04-01 | 2021-04-01 | Radar data filtering method based on thermodynamic diagram and graph theory clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113011520B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113432709A (en) * | 2021-06-25 | 2021-09-24 | 湖南工业大学 | Visualization mechanical fault diagnosis method based on graphics |
| CN114859300A (en) * | 2022-07-07 | 2022-08-05 | 中国人民解放军国防科技大学 | Radar radiation source data stream processing method based on graph connectivity |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050474A (en) * | 2014-06-10 | 2014-09-17 | 上海海洋大学 | Method for automatically extracting island shoreline based on LiDAR data |
| CN110766005A (en) * | 2019-10-23 | 2020-02-07 | 森思泰克河北科技有限公司 | Target feature extraction method and device and terminal equipment |
| CN111398943A (en) * | 2020-04-02 | 2020-07-10 | 森思泰克河北科技有限公司 | Target posture determining method and terminal equipment |
| CN111881930A (en) * | 2020-06-09 | 2020-11-03 | 广州市城市规划勘测设计研究院 | Thermodynamic diagram generation method and device, storage medium and equipment |
-
2021
- 2021-04-01 CN CN202110355748.3A patent/CN113011520B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050474A (en) * | 2014-06-10 | 2014-09-17 | 上海海洋大学 | Method for automatically extracting island shoreline based on LiDAR data |
| CN110766005A (en) * | 2019-10-23 | 2020-02-07 | 森思泰克河北科技有限公司 | Target feature extraction method and device and terminal equipment |
| CN111398943A (en) * | 2020-04-02 | 2020-07-10 | 森思泰克河北科技有限公司 | Target posture determining method and terminal equipment |
| CN111881930A (en) * | 2020-06-09 | 2020-11-03 | 广州市城市规划勘测设计研究院 | Thermodynamic diagram generation method and device, storage medium and equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113432709A (en) * | 2021-06-25 | 2021-09-24 | 湖南工业大学 | Visualization mechanical fault diagnosis method based on graphics |
| CN113432709B (en) * | 2021-06-25 | 2023-08-08 | 湖南工业大学 | Visual mechanical fault diagnosis method based on graphics |
| CN114859300A (en) * | 2022-07-07 | 2022-08-05 | 中国人民解放军国防科技大学 | Radar radiation source data stream processing method based on graph connectivity |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113011520B (en) | 2022-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110223302B (en) | Ship multi-target detection method based on rotation region extraction | |
| CN111192284B (en) | Vehicle-mounted laser point cloud segmentation method and system | |
| CN113011520B (en) | Radar data filtering method based on thermodynamic diagram and graph theory clustering | |
| CN104298998B (en) | A kind of data processing method of 3D point cloud | |
| CN107274417B (en) | Single tree segmentation method based on airborne laser point cloud aggregation relation | |
| CN107273902B (en) | A method of pylon point cloud is automatically extracted from on-board LiDAR data | |
| CN109325935A (en) | A kind of transmission line faultlocating method based on unmanned plane image | |
| CN115631356B (en) | Method and device for identifying missing of road facility, storage medium and electronic device | |
| CN109614936A (en) | The layered recognition method of remote sensing images Aircraft Targets | |
| CN103699903A (en) | City roof green area calculation method and system based on image identification | |
| CN104020510A (en) | Meteorological analysis method and device | |
| CN116520353B (en) | Ground detection method, device, storage medium and equipment based on laser point cloud | |
| CN115331190A (en) | Road hidden danger identification method and device based on radar fusion | |
| CN111624607A (en) | Low-altitude wind shear area acquisition method, device, equipment and storage medium | |
| CN105679103A (en) | Method for assessing air traffic volume accommodated in airport environment | |
| CN111488808A (en) | Lane line detection method based on traffic violation image data | |
| CN112597926A (en) | Method, device and storage medium for identifying airplane target based on FOD image | |
| CN110276318A (en) | Nighttime road rains recognition methods, device, computer equipment and storage medium | |
| KR20120119750A (en) | Method for identifying reflectivity cells associated with severe weather | |
| KR20120119749A (en) | Method for tracking reflectivity cells associated with severe weather | |
| CN108345007B (en) | Obstacle identification method and device | |
| CN111597939B (en) | High-speed rail line nest defect detection method based on deep learning | |
| CN110059544B (en) | Pedestrian detection method and system based on road scene | |
| CN109389152B (en) | Refined identification method for power transmission line falling object | |
| CN112924974B (en) | Method and device for identifying cloud cluster by using DBSCAN clustering algorithm and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |