CN115792825A - Method for inhibiting false radar tracks in traffic environment - Google Patents
Method for inhibiting false radar tracks in traffic environment Download PDFInfo
- Publication number
- CN115792825A CN115792825A CN202211312161.5A CN202211312161A CN115792825A CN 115792825 A CN115792825 A CN 115792825A CN 202211312161 A CN202211312161 A CN 202211312161A CN 115792825 A CN115792825 A CN 115792825A
- Authority
- CN
- China
- Prior art keywords
- track
- stable
- correlation
- flight
- value
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a false track suppression method in traffic environment, which comprises the steps of firstly, counting and calculating the characteristic variable correlation threshold of a false track and a real track through prior track data information; then judging the correlation between the temporary voyage and the formed determined stable voyage through multi-frame accumulation; then, judging whether the temporary flight is converted into an explicit stable flight or a recessive flight path according to the correlation, and if the temporary flight is converted into the recessive flight path, continuously judging the correlation between the recessive flight path and the stable flight determination through multi-frame accumulation; and finally, judging whether the implicit track is deleted or not according to the variation trend of the implicit stable range difference and whether the number of related frames is judged. The invention effectively solves the false track problem caused by large vehicle density and vehicle type difference in the traffic environment by utilizing the characteristic that the characteristic similarity of the same target track cannot be maintained and the characteristic difference of different target tracks is accumulated by a multi-frame accumulation method; the method can be applied to the field of radar track processing.
Description
Technical Field
The invention belongs to the technical field of radar data processing, and particularly relates to a method for inhibiting false tracks of a radar in a traffic environment.
Background
At present, millimeter wave radars based on a continuous wave system are more and more widely applied in the traffic field due to the characteristics of high resolution, all-weather working, small volume, reasonable price and the like. The method can realize the real-time positioning and tracking of multiple lanes and multiple targets in a high-density scene, can accurately count road condition vehicle information such as the average speed of lane vehicles, the traffic flow of lanes, the occupation rate of lanes, the queuing length of lanes and the like according to radar data, and can realize vehicle identification and classification even when the radar data has higher resolution.
An important premise for realizing the functions is that the radar can stably track dense vehicle targets in real time, namely, the dense vehicle targets have correct track data, but the millimeter wave radar has high resolution, and one target still has a plurality of echo point tracks after being processed by CFAR (computational fluid dynamics), under the condition, the number of the point tracks and the number of the track tracks are not in one-to-one correspondence, the correct matching of a plurality of the point tracks and a certain track is required, otherwise, a plurality of batches of false tracks are easily generated due to a considerable number of the point tracks.
In order to prevent the above phenomenon and reduce the subsequent processing pressure, the trace data needs to be condensed before the track processing. The current commonly used agglomeration algorithms mainly include a sliding window method and a clustering method, and the two algorithms are used for giving an agglomeration threshold, and points in the threshold are considered to be from the same target. However, vehicles on roads are various in types and different in shape, a large truck and a large trailer are more than 20 meters long, a common car is only 5-6 meters, in radar echo points, the points of the large car are relatively split, the span between the points is large, and often exceeds the length of the common car and the distance between the cars, so that a challenge is brought to point condensation threshold selection, the threshold selection is too large, target echo points which run in parallel or in front and back and have similar motion states are easily combined into one, the disappearance of real target point tracks is caused, and the outage or the loss of tracks is easily caused in the subsequent track processing process; too small a threshold selection will cause too heavy burden of subsequent processing and too many batches of false tracks, and this problem has not been completely solved so far.
An important link in the flight path processing process is data association, in the existing data association algorithm, a joint probability data interconnection algorithm (JPDA) proposed by a Bar-Shalom team and a multi-hypothesis tracking Method (MHT) proposed by Reid can achieve a good effect on tracking dense targets, but the calculation amount of the two algorithms is too large, in the engineering implementation, due to the limitation of low cost and small volume, the calculation capacity and storage space of a flight path processing platform selected by the existing millimeter wave radar are limited, and the small-sized computing platform is not suitable for the two algorithms, so that the existing association algorithm generally selects a nearest neighbor method (NN) with small calculation amount and a variant thereof, but the association algorithm can not solve the problem of multiple false batches of flight paths generated by a large vehicle.
In summary, the current track processing algorithm suitable for the millimeter wave radar cannot well solve the problem of multiple batches of false tracks caused by excessive trace points and dispersion.
Disclosure of Invention
The invention aims to provide a method for inhibiting false tracks of radars in a traffic environment, which aims to solve the technical problems that: considering that each confirmed stable track is originally converted from a temporary track after the temporary track meets the track starting condition, a plurality of batches of false tracks from the same target are not exceptional, and whether the temporary flight is related to a certain confirmed stable flight or not is judged by adopting the frame at the beginning of the temporary flight track, so that the accuracy is limited.
The technical scheme adopted by the invention is as follows:
the invention relates to a false track suppression method in traffic environment, which comprises the following steps:
s1, determining a characteristic variable correlation threshold value of the false track and the real track.
And S2, judging the correlation between the formed fixed stable flight and the temporary flight from the establishment of the temporary flight to the meeting of the flight path starting condition according to the characteristic variable correlation threshold.
And S3, judging whether the temporary voyage can be started normally as a dominant stable track or not according to the correlation. If yes, step S6 is executed, otherwise step S4 is executed.
And S4, converting the temporary flight into a hidden flight path, and continuously judging the correlation degree between the hidden flight path and the stable flight determination in the period of several weeks in the following preset frame.
And S5, judging whether the recessive flight path can be converted into a dominant stable flight path according to the correlation degree. If yes, step S6 is executed, otherwise step S7 is executed.
And S6, converting the temporary flight or the recessive flight path into an explicit stable flight path.
And S7, deleting the recessive track.
As a further scheme of the invention: the step S1 comprises the following steps:
s11, calculating a difference absolute value set of characteristic data S = { X, Y, V } of a plurality of tracks from the same target in the M frames according to the current prior track data informationWherein: x and Y are respectively the horizontal and vertical coordinate values of the target in Cartesian coordinates, V is the radial velocity,andthe absolute values of the differences of X, Y and V between the plurality of tracks of the k-th frame target t, respectively.
S12, after a plurality of frames are passed, arranging the absolute difference values of all the appeared feature data from small to large respectively, counting the number of the appeared values, establishing a distribution histogram, calculating the appearance probability of each value, selecting the absolute difference value of the corresponding feature data as the selected value of each feature variable threshold when the cumulative distribution function value is 90%, and obtaining a threshold set G thres = { deltax, deltay, deltav }, and the set of threshold values are used as thresholds associated with characteristic variables of false track and real trackThe value is obtained.
As a further scheme of the invention: the step S2 comprises the following steps:
S22, traversing all formed stable sailsCalculating the absolute value of the difference between the new temporary flight characteristic variable and the calculated value
S23, when the absolute value of the difference valueWhen the value is not more than the related threshold value of the characteristic variable, taking the corresponding formed stable navigation as the temporary stable navigation of the new temporary navigation, adding 1 to the value of the related frame number, and simultaneously carrying out normalization processing on the absolute value of the difference to obtain a normalized difference value
in the formula of lambda 1 And λ 2 Are not more than 1, respectively are characteristic value difference valuesAndthe proportion of the correlation in the overall correlation.
As a further scheme of the invention: in step S23, when the absolute value of the difference is smallerWhen the value in the parameter is larger than the related threshold value of the characteristic variable, a new temporary flight is re-established in the free point track of the next frame for judgment until the new temporary flight reaches the track starting condition.
As a further scheme of the invention: in step S23, the normalization formula is as follows:
as a further scheme of the invention: the step S3 comprises the following steps:
and S31, when the temporary flight reaches the track starting condition, counting the sum of the relevant frame numbers of all stable flights relevant to the temporary flight and the sum of the comprehensive relevance.
And S32, judging whether the sum of the stable navigation related frame numbers reaches a preset frame number value and whether the sum of the correlation degrees reaches a preset correlation value, if so, converting the sum of the stable navigation related frame numbers into an invisible flight path for management. Otherwise, the temporary flight is converted into a dominant normal stable flight path for tracking management.
As a further scheme of the invention: the preset frame value and the preset correlation value are related to the characteristics of the original point trace, when the front-end processing algorithm and the radar front-end parameter change, the characteristics of the original point trace change, and the preset frame value and the preset correlation value are correspondingly enlarged or reduced.
As a further scheme of the invention: in step S23: if a plurality of stable sails exist, the method meets the requirementsAnd selecting two stable navigations with the highest comprehensive correlation degree as the related stable navigations of the temporary navigations, and adding 1 to the number of the related frames respectively.
As a further scheme of the invention: step S4 comprises the following steps:
after several weeks of the following preset frames, judging whether the characteristic difference between the recessive flight path and a certain stable flight path has an increasing trend or not, or whether the number of frames without correlation is more than half of the period of the preset frames.
As a further scheme of the invention: in step S5, when the characteristic difference between the invisible track and a certain stable flight is increased, or the number of frames without correlation is larger than half of the preset frame number period, judging that the track is from a real target, and converting the invisible track into a dominant stable track for management.
The invention has the beneficial effects that: the invention provides a false track suppression method based on multi-frame accumulation in a traffic environment, which comprises the steps of firstly, counting and calculating a characteristic variable correlation threshold of a false track and a real track according to prior track data information; then setting the correlation threshold as a preset correlation threshold, and judging the correlation between the temporary navigation and the formed determined stable navigation through multi-frame accumulation; then, judging whether the transient flight is converted into a dominant stable flight or a recessive flight path according to the correlation; if the hidden flight path exists, continuously judging the relevance of the hidden flight path and the stable flight determination through multi-frame accumulation; and finally, judging whether the recessive track is eliminated or not according to the difference change trend of the two and the number of irrelevant frames. The invention utilizes the characteristics that the same target track features are similar and different target features are irrelevant, and effectively solves the false track problem caused by large vehicle density and vehicle type difference in the traffic environment by a multi-frame accumulation method of the difference between tracks.
Drawings
The invention will be further described with reference to the accompanying drawings.
Fig. 1 is a schematic view of a traffic radar installation.
FIG. 2 is a flow chart of a method for false track suppression in a traffic environment.
FIG. 3 logic diagram of a false track suppression method in a traffic environment
Fig. 4 is a flowchart of transient stationary navigation correlation determination.
FIG. 5 is a graph of the results of using a normal track processing algorithm.
FIG. 6 is a diagram of the results of the track processing implementation after the method is added.
Detailed Description
The invention is further described in conjunction with the drawings and the specific embodiments in order to facilitate understanding of the technical content of the invention by those skilled in the art.
In all embodiments of the present invention, the radar is installed at the intersection in a side-mounted manner, and is used for real-time monitoring and tracking of the vehicle target in the front 30m-250m section, as shown in fig. 1. Because the radar is installed with the height of about 7m, the radar has a blind area of about 30m, and 250m is the farthest irradiation range with stable radar points. Fig. 2 is a flow chart of a method for suppressing a false track in a traffic environment, and according to fig. 2, the method for suppressing the false track in the traffic environment of the present invention includes the following steps:
s1: and determining the correlation threshold value of the characteristic variables of the false track and the real track.
Wherein, the step S1 comprises the following steps S11-S12.
S11, calculating a difference absolute value set of characteristic data S = { X, Y, V } of a plurality of tracks from the same target in the M frames according to the current prior track data informationWherein: x and Y are respectively the horizontal and vertical coordinate values of the target under Cartesian coordinates, V is the radial velocity,andthe absolute values of the differences of X, Y and V between the plurality of tracks of the kth frame target t, respectively.
Specifically, in implementation, 1000 frames of track data information is selected according to a currently used track processing algorithm, and a difference set of information such as an X coordinate, a Y coordinate, a speed and the like of each feature S = { X, Y, V } between multiple tracks from the same target in the period is countedAnd stored.
S12, after a plurality of frames are passed, arranging the absolute difference values of all the appeared feature data from small to large respectively, counting the number of the appeared values, establishing a distribution histogram, calculating the appearance probability of each value, selecting the absolute difference value of the corresponding feature data as the selected value of each feature variable threshold when the cumulative distribution function value is 90%, and obtaining a threshold set G thres And (4) the set of threshold values is used as the characteristic variable correlation threshold values of the false track and the real track.
In specific implementation, after 1000 frames, the difference values of all the appeared characteristic values are respectively arranged from small to large, then the number of the appeared values is counted, a distribution histogram is established, the probability of the appeared values is calculated, the corresponding characteristic value difference value when the cumulative distribution function value is 90% is selected as the selected value of each characteristic variable threshold, and finally a threshold set G is obtained thres ={ΔX thres ,ΔY thres ,ΔV thres And taking the set of threshold values as a preset wave gate of subsequent transient stable navigation correlation.
S2: and judging the correlation between the formed stable navigation and the formed stable navigation during the period from the establishment of the temporary navigation to the meeting of the track starting condition according to the characteristic variable correlation threshold.
Step S2 comprises the following steps S21-S24, the logic diagram being shown in fig. 3.
S21, establishing a new temporary voyage by using the free point trace (namely the point trace which is not related to the stable voyage and the established temporary voyage)
S22, traversing all formed stable navigationsCalculating the absolute value of the difference between the new temporary flight characteristic variable and the calculated value
S23, when the absolute value of the difference valueWhen the value is not more than the related threshold value of the characteristic variable, taking the corresponding formed stable navigation as the temporary stable navigation of the new temporary navigation, adding 1 to the value of the related frame number, and simultaneously carrying out normalization processing on the absolute value of the difference to obtain a normalized difference value
The normalization method comprises the following steps:
it should be noted that, first, when the absolute value of the difference is smaller than the threshold valueWhen the value in the parameter is larger than the characteristic variable correlation threshold value, a new temporary flight is reestablished in the free point track of the next frame for judgment until the new temporary flight reaches the flight track starting condition. That is, steps S21 to S23 are repeatedly executed until a tentative flight reaches the track start condition.
Secondly, if a plurality of stable sails exist, the requirement of stable sailing is metAnd selecting two stable navigations with the highest comprehensive correlation degree as the related stable navigations of the temporary navigations, and adding 1 to the number of the related frames respectively.
Wherein λ is 1 And λ 2 Are not more than 1, respectively are characteristic value difference valuesAndthe proportion of the correlation in the overall correlation. When in some specific embodiments, λ 1 And λ 2 The value of (b) is generally set to 0.3.
S3: and judging whether the temporary voyage can be started normally as the dominant stable track or not according to the correlation. If yes, step S6 is executed, otherwise step S4 is executed.
Referring to FIG. 4, step S3 includes the following steps S31-S32.
And S31, when a temporary flight reaches a flight path starting condition, counting the sum of the relevant frame numbers of all stable flights relevant to the temporary flight and the sum of the comprehensive relevance.
In some embodiments of the present invention, when a transient voyage reaches a track start condition, a 6/8 logic start method based on a sliding window is selected, so that M =8 frames are required for setting the track start, and the sum of the number of relevant frames of all stable voyages related to the transient voyage and the sum of the comprehensive correlation degree in the 8 frames are counted.
And S32, judging whether the sum of the stable navigation related frame numbers reaches a preset frame number value and whether the sum of the correlation degrees reaches a preset correlation value, if so, converting the sum of the stable navigation related frame numbers into an invisible flight path for management. Otherwise, the temporary flight is converted into an explicit normal stable flight path for tracking management.
In a corresponding embodiment, if the sum of the number of the stable navigation related frames reaches a preset frame number value and the sum of the correlation degree reaches a preset related value, preliminarily considering that the temporary navigation may come from a large-scale target multi-batch false track, converting the temporary navigation into an invisible track for management, and sending information of the temporary navigation to a field ViewFlag =0 of a subsequent display interface, wherein data information of the invisible track does not participate in analysis of functions such as traffic flow statistics, average vehicle speed measurement and calculation and the like; and if the judgment result is that the temporary navigation does not have strong correlation with any stable navigation, setting the field ViewFlag to be 1, and converting the temporary navigation into an explicit normal stable flight path for tracking management.
It should be noted that the preset frame number value and the preset correlation value are related to the characteristics of the original point trace, when the front-end processing algorithm and the radar front-end parameter change, the characteristics of the original point trace change, and the preset frame number value and the preset correlation value are correspondingly enlarged or reduced.
And S4, converting the temporary flight into a hidden flight path, and continuously judging the correlation degree between the hidden flight path and the stable flight determination in the period of several weeks in the following preset frame.
Specifically, after several weeks of the following preset frames, whether the characteristic difference between the recessive flight path and a certain stable flight path has an increasing trend or whether the number of frames without correlation is greater than a half of the preset frame period is judged.
And S5, judging whether the recessive flight path can be converted into a dominant stable flight path according to the correlation degree. If yes, step S6 is executed, otherwise step S7 is executed.
Specifically, when the characteristic difference between the recessive flight path and a certain stable flight path is increased, or the number of frames without correlation is greater than half of the preset frame number period, it is determined that the flight path is from a real target, the field ViewFlag is set to 1, and the recessive flight path is converted into the dominant stable flight path for management. Otherwise, judging that the two are correlated, wherein the probability of the two being from the same target is higher, and the invisible flight path is a plurality of false flight paths of a certain target and is deleted.
And S6, converting the temporary flight or the recessive flight path into an explicit stable flight path.
And S7, deleting the hidden track.
Referring to fig. 5 and 6, fig. 5 and 6 are graphs comparing the normal track processing and the track processing implementation result after the method is added, respectively, and it can be seen from the comparison that after the method is added to the track processing algorithm, the false tracks from the same target are suppressed.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A false track suppression method in a traffic environment is characterized by comprising the following steps:
s1, determining a characteristic variable correlation threshold value of a false track and a real track;
s2, judging the correlation between the formed fixed stable flight and the temporary flight from the establishment of the temporary flight to the meeting of the flight path starting condition according to the characteristic variable correlation threshold;
s3, judging whether the temporary voyage can be started normally as a dominant stable track or not according to the correlation; if yes, executing step S6, otherwise executing step S4;
s4, converting the temporary flight into a hidden flight path, and continuously judging the correlation degree between the hidden flight path and the stable flight determination within a plurality of weeks of a subsequent preset frame;
s5, judging whether the recessive flight path can be converted into a dominant stable flight path according to the correlation degree; if yes, executing step S6, otherwise, executing step S7;
s6, converting the temporary flight or the recessive flight path into a dominant stable flight path;
and S7, deleting the hidden track.
2. The method for suppressing the false track in the traffic environment according to claim 1, wherein step S1 includes:
s11, calculating a difference absolute value set of characteristic data S = { X, Y, V } of a plurality of tracks from the same target in the M frames according to the current prior track data informationWherein: x and Y are respectively the horizontal and vertical coordinate values of the target under Cartesian coordinates, V is the radial velocity,andabsolute values of differences of X, Y and V among a plurality of tracks of the k frame target t are respectively;
s12, after several frames, respectively arranging the absolute difference values of all the appeared characteristic data from small to large, counting the number of the appeared values, establishing a distribution histogram, calculating the appearance probability of each value, selecting the absolute difference value of the corresponding characteristic data as the selected value of each characteristic variable threshold when the cumulative distribution function value is 90%, and obtaining a threshold set G thres = { Δ X, Δ Y, Δ V }, and this set of thresholds is used as the characteristic variable correlation thresholds for false track and true track.
3. The method for suppressing the false tracks in the traffic environment according to claim 1, wherein the step S2 comprises:
S22, traversing all formed stable navigationsCalculating the absolute value of the difference between the new temporary flight characteristic variable and the calculated value
S23, when the absolute value of the difference valueWhen the value is not more than the related threshold value of the characteristic variable, the corresponding formed stable navigation is taken as the temporary stable navigation of the new temporary navigation, the value of the related frame number is added with 1, and the absolute value of the difference is returned at the same timeNormalizing to obtain normalized difference
4. The method as claimed in claim 3, wherein the step S23 is performed when the absolute value of the difference is smaller than the thresholdWhen the value in the parameter is larger than the related threshold value of the characteristic variable, a new temporary flight is re-established in the free point track of the next frame for judgment until the new temporary flight reaches the track starting condition.
6. the method for suppressing false tracks in traffic environment according to claim 3, wherein step S3 comprises:
s31, when a temporary flight reaches a flight path starting condition, counting the sum of all stable flight related frame numbers related to the temporary flight and the sum of comprehensive correlation degrees;
s32, judging whether the sum of the stable navigation related frame numbers reaches a preset frame number value and whether the sum of the correlation degrees reaches a preset correlation value, if so, converting the sum of the stable navigation related frame numbers into an invisible flight path for management; otherwise, the temporary flight is converted into a dominant normal stable flight path for tracking management.
7. The method according to claim 6, wherein the preset frame number and the preset correlation value are related to the characteristics of the original point trace, when the front-end processing algorithm and the radar front-end parameter change, the characteristics of the original point trace change, and the preset frame number and the preset correlation value are correspondingly enlarged or reduced.
8. The method for suppressing false tracks in traffic environment according to claim 3, wherein in step S23: if a plurality of stable sails exist, the method meets the requirementsAnd selecting two stable navigations with the highest comprehensive correlation degree as the related stable navigations of the temporary navigations, and adding 1 to the number of the related frames respectively.
9. The method for suppressing false tracks in traffic environment according to claim 1, wherein step S4 includes:
after several weeks of the following preset frames, judging whether the characteristic difference between the recessive flight path and a certain stable flight path has an increasing trend or not, or whether the number of frames without correlation is more than half of the period of the preset frames.
10. The method according to claim 9, wherein in step S5, when the characteristic difference between the hidden track and a certain stable track is increasing or the number of frames without correlation is greater than half of the preset frame number period, it is determined that the track is from a real target, and the hidden track is changed to a dominant stable track for management.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211312161.5A CN115792825A (en) | 2022-10-25 | 2022-10-25 | Method for inhibiting false radar tracks in traffic environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211312161.5A CN115792825A (en) | 2022-10-25 | 2022-10-25 | Method for inhibiting false radar tracks in traffic environment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115792825A true CN115792825A (en) | 2023-03-14 |
Family
ID=85433691
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211312161.5A Pending CN115792825A (en) | 2022-10-25 | 2022-10-25 | Method for inhibiting false radar tracks in traffic environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115792825A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117092636A (en) * | 2023-10-12 | 2023-11-21 | 深圳安智杰科技有限公司 | System and method for recognizing false track of millimeter wave radar multi-target tracking |
-
2022
- 2022-10-25 CN CN202211312161.5A patent/CN115792825A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117092636A (en) * | 2023-10-12 | 2023-11-21 | 深圳安智杰科技有限公司 | System and method for recognizing false track of millimeter wave radar multi-target tracking |
CN117092636B (en) * | 2023-10-12 | 2024-01-02 | 深圳安智杰科技有限公司 | System and method for recognizing false track of millimeter wave radar multi-target tracking |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111540201B (en) | Vehicle queuing length real-time estimation method and system based on roadside laser radar | |
CN110542885B (en) | Millimeter wave radar target tracking method in complex traffic environment | |
CN113671481B (en) | 3D multi-target tracking processing method based on millimeter wave radar | |
CN113253257B (en) | Strip mine obstacle detection method based on multi-millimeter-wave radar and vision | |
CN115792825A (en) | Method for inhibiting false radar tracks in traffic environment | |
CN112731296A (en) | Automobile millimeter wave radar point trace condensation method and system | |
CN113370977A (en) | Intelligent vehicle forward collision early warning method and system based on vision | |
CN114898326A (en) | Method, system and equipment for detecting reverse running of one-way vehicle based on deep learning | |
CN114879192A (en) | Decision tree vehicle type classification method based on road side millimeter wave radar and electronic equipment | |
CN106405539A (en) | Vehicle radar system and method for removing non-interested target | |
CN110648542B (en) | High-precision vehicle flow detection system based on azimuth recognition narrow-wave radar | |
CN115561750A (en) | Millimeter wave radar vehicle type classification method | |
CN115932831A (en) | Target segment tracking method, device, equipment and storage medium based on radar | |
CN112101443B (en) | Small group track starting method based on measurement processing under multi-group target scene | |
CN115980738B (en) | Multi-target tracking method based on dynamic and static separation | |
Zhong et al. | Research on Road Object Detection Algorithm Based on YOLOv5+ Deepsort | |
CN115062683A (en) | 4D millimeter wave radar clustering method based on DBSCAN and storage medium | |
CN114670851A (en) | Driving assistance system, method, terminal and medium based on optimizing tracking algorithm | |
CN113012430A (en) | Vehicle queuing length detection method, device, equipment and readable storage medium | |
CN114049610A (en) | Active discovery method for reverse and reverse illegal behaviors of motor vehicles on expressway | |
CN114387310A (en) | Urban trunk road traffic flow statistical method based on deep learning | |
CN114633782B (en) | Train arrival early warning method, device, equipment and medium for railway level crossing | |
WO2023108930A1 (en) | Point cloud speed-based millimeter wave radar big car identification method | |
Zhao et al. | Millimeter wave radar denoising and obstacle detection in highly dynamic railway environment | |
CN114882699B (en) | Road section traffic risk identification method based on conflict space-time correlation characteristics in area |
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 |