CN107633681B - Method for detecting abnormal event by using Doppler radar - Google Patents
Method for detecting abnormal event by using Doppler radar Download PDFInfo
- Publication number
- CN107633681B CN107633681B CN201710518811.4A CN201710518811A CN107633681B CN 107633681 B CN107633681 B CN 107633681B CN 201710518811 A CN201710518811 A CN 201710518811A CN 107633681 B CN107633681 B CN 107633681B
- Authority
- CN
- China
- Prior art keywords
- radar
- speeds
- doppler radar
- overspeed
- state transition
- 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.)
- Active
Links
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method for detecting abnormal events by using a Doppler radar, which comprises the following steps: installing a Doppler radar beside the camera rack, monitoring a road by using the Doppler radar, and acquiring radar data; calculating a transition probability function between every two speeds, and acquiring a random process; constructing a state transition network by collecting radar data of a vehicle under a normal condition within a period of time; searching a state transition network, and calculating two adjacent time speeds A in a random process1Is transferred to speed A3The probability of the state transition is obtained, so that the most probable situation is obtained, namely, the early warning is given. The method for detecting the abnormal events by using the Doppler radar can assist the camera to carry out intelligent event detection under adverse conditions, can effectively detect the events of abnormal parking, overspeed, slow speed and reverse running of the road vehicle in real time, and has high detection efficiency and application prospect.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to a method for detecting abnormal events by using a Doppler radar.
Background
In recent years, with the increasing amount of traffic, the problem of traffic congestion has become increasingly prominent. The traffic incidents (including traffic accidents) on the road are gradually increased, which brings great inconvenience to people's daily trips and even threatens people's lives and properties. Meanwhile, traffic accidents also reduce the transportation efficiency of roads, the average speed of vehicles is reduced, and the operation of roads is seriously influenced. Therefore, once a traffic event occurs on a road, it is necessary to quickly find the event and respond, and to reduce the damage caused by the event as much as possible. The traffic incident detection means that various detection methods are used for distinguishing traffic incidents on roads, and is an important content in an intelligent transportation system. At present, video is mainly used for detecting abnormal events such as overspeed, retrograde motion, pedestrian, parking and the like in the event detection in the field of intelligent traffic. The video detection range is wide, various events can be detected, the cost is low, the maintenance is simple, the camera can have the situation of large error or even incapability due to the interference of adverse weather, light change, camera shaking and the like, and great difficulty can be caused to the rule judgment in the later period. Based on the above statements, the present invention proposes a method for detecting abnormal events using doppler radar.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a method for detecting abnormal events by using a Doppler radar.
A method of detecting abnormal events using doppler radar comprising the steps of:
s1, mounting a Doppler radar beside the camera rack, and monitoring the road by using the Doppler radar to acquire radar data;
s2, converting and acquiring speeds at corresponding moments according to the radar data acquired in the step S1, determining corresponding relations between speeds at different moments by calculating transition probability functions of every two speeds of the same vehicle at different moments, and decomposing different speeds of different vehicles at the same moment into independent random processes respectively;
s3, constructing a state transition network by collecting radar data of a vehicle under normal conditions in a period of time, wherein M (x, y) { P }x,y(stop) Px,y(overspeed), Px,y(Slow, Px,y(retrograde), … … } is a set containing abnormal event state probabilities;
s4, searching a state transfer network, and calculating the speeds A of two adjacent moments of the same vehicle in the random process1Is transferred to speed A3The state transition probability of (a), namely: m (A)1,A3)＝{PA1,A2(parking) ∙ PA2,A3(stop) PA1,A2(overspeed) ∙ PA2,A3(overspeed), PA1,A2(Slow) ∙ PA2,A3(Slow, PA1,A2(retrograde) ∙ PA2,A3(reverse), … … }, so as to obtain the most probable situation, namely giving an early warning.
Preferably, the installation angle of the doppler radar is the same as the installation angle of the camera in step S1.
Preferably, the radar data in step S1 is in a range of-128 to 127, where positive numbers represent that the moving direction of the object is close to the radar, negative numbers represent that the moving direction of the object is far from the radar, and the radar data has a one-to-one correspondence relationship with the speed.
Preferably, the calculation formula of the transition probability in step S2 is as follows:
where Z is the normalization factor and α is the adjustment factor.
Preferably, the abnormal events in step S3 include parking, speeding, slowing and reversing events.
The method for detecting the abnormal events by using the Doppler radar effectively overcomes the defects that video detection is not limited by environment and is influenced by interference of weather, illumination, shaking and the like in the prior art by using the Doppler radar for detection, can assist a camera to carry out intelligent event detection under adverse conditions, can effectively detect the abnormal parking, overspeed, slow speed and retrograde motion events of the road vehicle in real time, and has high detection efficiency and application prospect.
Drawings
Fig. 1 is a schematic diagram illustrating an output of a doppler radar in step S1 according to an embodiment of the method for detecting an abnormal event by using the doppler radar in the present invention;
fig. 2 is a schematic diagram illustrating radar speeds at different times in step S2 according to an embodiment of the method for detecting an abnormal event by using a doppler radar according to the present invention;
fig. 3 is a schematic diagram illustrating a transition state of radar speed at different times in step S2 according to an embodiment of the method for detecting an abnormal event by using a doppler radar according to the present invention;
FIG. 4 is a schematic diagram of two independent random processes in step S2 according to an embodiment of the method for detecting abnormal events by using Doppler radar;
fig. 5 is a schematic diagram of a local state transition network in step S3 according to an embodiment of the method for detecting an abnormal event by using a doppler radar.
Detailed Description
The invention will be further illustrated with reference to specific embodiments, with reference to fig. 1-5.
Examples
The invention provides a method for detecting abnormal events by using a Doppler radar, which comprises the following steps:
s1, installing a doppler radar beside the camera rack, monitoring the road by the doppler radar, and acquiring radar data, as shown in fig. 1: the abscissa ranges from-128 to 127, 256 discrete values, which represent Fast Fourier Transform (FFT) values at different frequencies, where different velocities are measured, the range of 0 to 128 is measured, the positive number represents that the moving direction of the object is close to the radar, the negative number represents that the moving direction of the object is far from the radar, B represents a value of FFT, A represents a corresponding threshold, if the value of FFT exceeds the threshold, it represents that there is a linear relationship between the abscissa corresponding to the point and the velocity of the moving object, for the sake of convenience, the subscript value and velocity are in a one-to-one correspondence, i.e., 0 represents that the velocity is 0, 127 represents that the moving object is close to the radar at a velocity of km 127/h, -128 represents that the moving object is far from the radar at a velocity of 128km/h, the output of the radar is regarded as two sets of one-bit arrays containing 128 items of data, the subscript of the arrays is 0 to 127, these two arrays are now represented as A0 to A127 (1 to 128 representing, b0 to B127 (0 to 127 representing the right half);
s2, converting the radar data obtained in step S1 to obtain the speed of the corresponding time, assuming that the radar outputs 15 sets of data per second, as shown in fig. 2, in order to determine the corresponding relationship between the speeds at different times, assuming that the speeds of the moving object at different times conform to the markov random process, i.e. the state of the system at the next time is determined only by the current state, and is not dependent on any previous state, then t is t0If A m in a group of time data]If the term is 1, t is t0+1 time A [ n]The transition probability with term 1 is calculated by the following formula:
in formula (1), Z is a normalization factor, and α is an adjustmentFactor, let t0At a moment of time v1、v2、v3Three speeds, t1At a moment of time v4、v5、v6Three speeds, the corresponding relation between the speeds at different moments can be determined with the maximum possibility by calculating the transition probability function between every two speeds, the obtained result is shown in fig. 3, and a plurality of speeds at the same moment are decomposed into independent random processes on a time scale through a step S2, as shown in fig. 4, so that the subsequent processing is facilitated;
s3, a state transition network is constructed by collecting radar data of a vehicle under abnormal conditions such as stopping, speeding, slowing and reversing under different conditions within a period of time, namely M (x, y) ═ Px,y(stop) Px,y(overspeed), Px,y(Slow, Px,y(retrograde), … … } is a set containing abnormal event state probabilities, as shown in FIG. 5;
s4, searching the state transition network, calculating the state transition probability of the speed A [31] to the speed A [34] at two adjacent moments in the random process, obtaining two shortest paths from the state transition network, namely A [31] → A [32] → A [34] and A [31] → A [32] → A [34], and respectively calculating the state probabilities of the two paths, namely:
M(31,34)＝{P31,32(Normal). P32,34(Normal), P31,32(overspeed). P32,34(overspeed), P31,32(parking). P32,34(stop) P31,32(retrograde). P32,34(go in the wrong direction) }
M(31,34)＝{P31,33(Normal). P33,34(Normal), P31,33(overspeed). P33,34(overspeed), P31,33(parking). P33,34(stop) P31,33(retrograde). P33,34(go in the wrong direction) }
And finding out the state with the highest probability from the two states as the state transition probability of M (31,34) so as to obtain the most probable situation, namely giving an early warning.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (5)
1. A method for detecting abnormal events using doppler radar, comprising the steps of:
s1, mounting a Doppler radar beside the camera rack, and monitoring the road by using the Doppler radar to acquire radar data;
s2, converting and acquiring speeds at corresponding moments according to the radar data acquired in the step S1, determining corresponding relations between speeds at different moments by calculating transition probability functions of every two speeds of the same vehicle at different moments, and decomposing different speeds of different vehicles at the same moment into independent random processes respectively;
s3, constructing a state transition network by collecting radar data of a vehicle under normal conditions in a period of time, wherein M (x, y) { P }x,y(stop) Px,y(overspeed), Px,y(Slow, Px,y(retrograde), … … } is a set containing abnormal event state probabilities;
s4, searching a state transfer network, and calculating the speeds A of two adjacent moments of the same vehicle in the random process1Is transferred to speed A3The state transition probability of (a), namely: m (A)1,A3)＝{PA1,A2(parking) ∙ PA2,A3(stop) PA1,A2(overspeed) ∙ PA2,A3(overspeed), PA1,A2(Slow) ∙ PA2,A3(Slow, PA1,A2(retrograde) ∙ PA2,A3(reverse), … … }, so as to obtain the most probable situation, namely giving an early warning.
2. The method of claim 1, wherein the installation angle of the doppler radar is the same as the installation angle of the camera in step S1.
3. The method of claim 1, wherein the radar data in step S1 is in a range of-128 to 127, wherein positive numbers represent that the object is moving closer to the radar and negative numbers represent that the object is moving farther from the radar, and the radar data has a one-to-one relationship with velocity.
4. The method of claim 1, wherein the transition probability in step S2 is calculated by the following formula:
where Z is the normalization factor and α is the adjustment factor.
5. The method of claim 1, wherein the abnormal events in step S3 include stop, overspeed, slow speed and reverse driving events.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710518811.4A CN107633681B (en) | 2017-06-30 | 2017-06-30 | Method for detecting abnormal event by using Doppler radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710518811.4A CN107633681B (en) | 2017-06-30 | 2017-06-30 | Method for detecting abnormal event by using Doppler radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107633681A CN107633681A (en) | 2018-01-26 |
| CN107633681B true CN107633681B (en) | 2020-05-29 |
Family
ID=61099263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710518811.4A Active CN107633681B (en) | 2017-06-30 | 2017-06-30 | Method for detecting abnormal event by using Doppler radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107633681B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20000033531A (en) * | 1998-11-24 | 2000-06-15 | 김영환 | Drive method of hidden markov model parameter recognizer using car speed adaptive data base |
| KR20040090228A (en) * | 2003-04-16 | 2004-10-22 | 윤태종 | System for confirming speed measurement of an overspeed-measuring camera |
| CN101324667A (en) * | 2007-06-13 | 2008-12-17 | 邹谋炎 | Design of vehicle velocity detection radar and signal processing method |
| CN101599219A (en) * | 2008-06-04 | 2009-12-09 | 新南威尔士州道路交通管理局 | Traffic signal control system |
| CN103448730A (en) * | 2013-09-17 | 2013-12-18 | 东南大学 | Method of estimating key alarm parameters in highway automobile rear-end collision |
| CN104408924A (en) * | 2014-12-04 | 2015-03-11 | 深圳北航新兴产业技术研究院 | Detection method for abnormal traffic flow of urban road based on coupled hidden markov model |
| CN106023601A (en) * | 2016-07-22 | 2016-10-12 | 池州学院 | Radar speed measurement control system |
-
2017
- 2017-06-30 CN CN201710518811.4A patent/CN107633681B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20000033531A (en) * | 1998-11-24 | 2000-06-15 | 김영환 | Drive method of hidden markov model parameter recognizer using car speed adaptive data base |
| KR20040090228A (en) * | 2003-04-16 | 2004-10-22 | 윤태종 | System for confirming speed measurement of an overspeed-measuring camera |
| CN101324667A (en) * | 2007-06-13 | 2008-12-17 | 邹谋炎 | Design of vehicle velocity detection radar and signal processing method |
| CN101599219A (en) * | 2008-06-04 | 2009-12-09 | 新南威尔士州道路交通管理局 | Traffic signal control system |
| CN103448730A (en) * | 2013-09-17 | 2013-12-18 | 东南大学 | Method of estimating key alarm parameters in highway automobile rear-end collision |
| CN104408924A (en) * | 2014-12-04 | 2015-03-11 | 深圳北航新兴产业技术研究院 | Detection method for abnormal traffic flow of urban road based on coupled hidden markov model |
| CN106023601A (en) * | 2016-07-22 | 2016-10-12 | 池州学院 | Radar speed measurement control system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107633681A (en) | 2018-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6571545B2 (en) | Object detection apparatus and object detection method | |
| CN104809878B (en) | Method for detecting abnormal condition of urban road traffic by utilizing GPS (Global Positioning System) data of public buses | |
| CN109285349B (en) | Method for detecting highway traffic incident under cooperative vehicle and road environment and early warning system | |
| CN103500503A (en) | Method and system for analyzing accurate road conditions based on crowdsourcing mode | |
| JP2017125802A (en) | Object detection device and object detection method | |
| CN111524365A (en) | Method for classifying vehicle types by using multiple geomagnetic sensors | |
| CN112562329A (en) | Traffic event sensing method based on millimeter wave radar data | |
| CN102923000B (en) | A kind of automobile based on binocular vision target detect is initiatively moveed backward speed-limiting control method | |
| CN111976638B (en) | Vehicle collision detection method and system based on vehicle-mounted intelligent terminal | |
| CN103578242A (en) | Private car driving monitoring and alarm system and method thereof | |
| CN109272760B (en) | Online detection method for abnormal data value of SCATS system detector | |
| CN107633681B (en) | Method for detecting abnormal event by using Doppler radar | |
| CN107657811B (en) | Road event detection method based on Doppler radar and Bayesian network | |
| CN117238145A (en) | Intelligent traffic management method and system based on big data | |
| KR100378956B1 (en) | Acoustical Vehicle Detection System | |
| CN111311928A (en) | Speed measuring system and speed measuring method of double geomagnetic sensors | |
| CN116935631A (en) | Abnormal traffic situation detection method, device and system based on radar fusion | |
| Liu et al. | Pattern recognition of vehicle types and reliability analysis of pneumatic tube test data under mixed traffic condition | |
| CN108806260A (en) | A kind of multiple target flow statistical method based on radar | |
| CN114596711A (en) | Accident responsibility determination method, device, equipment and storage medium | |
| CN113887659A (en) | Vehicle type classification algorithm based on naive Bayes | |
| CN108648479B (en) | Device and method for reminding night foggy road section in real time by using electronic map | |
| CN107833464B (en) | Driving behavior safety assessment method and storage medium | |
| CN107590998B (en) | Road node running state recognition system based on floating car data | |
| CN109367478B (en) | Automobile danger grading and early warning system and method |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Rao Xiaochun Inventor after: Yu Yang Inventor after: Zhang Yuxi Inventor after: Fang Zhengpeng Inventor before: Rao Xiaochun Inventor before: Yu Ze Inventor before: Zhang Yuxi Inventor before: Fang Zhengpeng |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |