CN113514848B - Railway crossing locomotive detecting system based on laser scanning - Google Patents
Railway crossing locomotive detecting system based on laser scanning Download PDFInfo
- Publication number
- CN113514848B CN113514848B CN202110354002.0A CN202110354002A CN113514848B CN 113514848 B CN113514848 B CN 113514848B CN 202110354002 A CN202110354002 A CN 202110354002A CN 113514848 B CN113514848 B CN 113514848B
- Authority
- CN
- China
- Prior art keywords
- locomotive
- area
- crossing
- data
- monitoring
- 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
- 230000003137 locomotive effect Effects 0.000 title claims abstract description 154
- 238000001514 detection method Methods 0.000 claims abstract description 89
- 238000012544 monitoring process Methods 0.000 claims description 65
- 238000000034 method Methods 0.000 claims description 20
- 238000005259 measurement Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 8
- 238000010586 diagram Methods 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000009825 accumulation Methods 0.000 claims description 2
- 238000013459 approach Methods 0.000 claims description 2
- 230000001174 ascending effect Effects 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims 1
- 230000006870 function Effects 0.000 abstract description 6
- 230000006399 behavior Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/04—Systems determining the presence of a target
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
- G01S17/08—Systems determining position data of a target for measuring distance only
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/50—Systems of measurement based on relative movement of target
- G01S17/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses a railway crossing locomotive detection system based on laser scanning, which comprises a system and a laser radar, wherein the system comprises the laser radar and a locomotive detection algorithm, the laser radar is installed in a railway crossing area, then the area detection range is divided, the railway crossing is scanned in real time to form dynamic data, crossing background self-learning algorithm and locomotive area detection algorithm are adopted to detect crossing locomotive objects, then targets in a scene are extracted from the real-time scanning data, the current scanning locomotive object targets are identified and classified, then a time state data sequence of the locomotive entering and exiting the crossing is established according to the scanning data, and finally the running speed of the locomotive is calculated. The invention effectively solves the problems of few locomotive detection methods, poor detection stability and imperfect functions of the current complex background of the railway crossing, provides support for the full-automatic safe operation of the unmanned railway crossing, and further improves the intelligent level of the safety supervision of the railway crossing.
Description
Technical Field
The invention relates to the technical field of intelligent supervision of port railway traffic, in particular to a railway crossing locomotive detection system based on laser scanning.
Background
Most detection systems of short-distance vehicle intrusion alarm detection technology of modern railway crossing in China have simple control logic, consume a large amount of manpower and material resources to participate in manual control in the whole process, and for the research of railway crossing detection, the earliest adopted method is a track circuit, and the track circuit has the characteristics of stable operation and simple installation because of more defects, such as single function, easy false alarm, and large installation task amount, and then an embedded monitoring system based on infrared sensing technology is developed and adopted, but the detection accuracy of the technology is limited, the final result is influenced because of the difference of the sensor heights, so the technology is not suitable for the practical application of railway crossing detection of the current industrial field, and the current method is more suitable for the practical application of the industrial field because of adopting an alarm detection system based on an axle counting sensor.
With the continuous monitoring use of alarm detection systems, the following problems are found during use:
1. the controllers of some existing alarm detection systems are relatively complex and have low reliability, the distance measurement of some detection systems can be influenced by other factors (such as frequency band loss and the like), and the detection of the detection systems is only limited to the relative distance and the relative speed of a measurement target object, so that the detected object cannot be accurately identified;
2. In addition, some existing detection systems are all passively receiving data, the perception effect is greatly influenced by environmental factors, and some key problems cannot be solved.
It is desirable to design a railroad grade crossing locomotive detection system based on laser scanning to address the above-mentioned problems.
Disclosure of Invention
The invention aims to provide a railway crossing locomotive detection system based on laser scanning, which aims to solve the problems that the controllers of the existing alarm detection systems are relatively complex and low in reliability, the distance measurement of part of detection systems can be influenced by other factors (such as frequency band loss and the like), and the detection of the detection systems is only limited to the relative distance and the relative speed of a measurement target object, so that detected objects cannot be accurately identified, the existing detection systems are all passive receiving data, the perception effect of the detection systems is greatly influenced by environmental factors, and key problems cannot be solved.
In order to achieve the above purpose, the present invention provides the following technical solutions: the system comprises the laser radar and the locomotive detection algorithm, and the locomotive detection method comprises the railway crossing detection data initialization, the locomotive object detection, the locomotive running direction discrimination mechanical locomotive speed calculation, wherein:
railway crossing detection data initialization
The method comprises the steps of carrying out custom drawing on a road junction laser radar scanning area, then carrying out adjustment and edge finishing on the shape of the area according to the on-site working condition of a railway road junction to obtain a required area outline, carrying out area division by adopting a 'depth range' expression mode, wherein main data of a data structure comprise a start angle and a stop angle of an area scanning angle, the depth range comprises a start depth and a stop depth, the range represents a ray range distance range of the area on the scanning angle, depth measurement data on the scanning angle only belong to the area when the depth measurement data are positioned in the range, after the geometric definition of an area group is completed, the area in the group is converted into a uniform internal data structure, adopting polar coordinate representation to improve the efficiency of an area monitoring algorithm, and further initializing a background outline by utilizing a road junction background self-learning and locomotive area monitoring algorithm, wherein the road junction background self-learning algorithm refers to the accumulation of real-time scanning data of a current scene for a period so as to detect the background outline of the current scene, and accordingly generating background data of area monitoring; the locomotive area monitoring algorithm is used for extracting all targets in a monitoring area group in a scene from real-time scanning data, classifying and identifying current monitoring targets according to operation control parameters, eliminating irrelevant targets and detecting effective monitoring targets;
Locomotive object detection
In order to scan the two-dimensional information of a real object, the laser radar needs to scan and detect the emitted laser beam in the horizontal direction to obtain laser point cloud data which has huge data, wide vision range, high resolution and certain measurement noise and other characteristics, because the locomotive has larger volume and other factors, the laser point cloud data obtained by the radar is more than other detection targets and data points are more dense, aiming at a railway crossing area needing to be detected, the laser point cloud characteristics of the target locomotive are observed, when the locomotive approaches, the number of the laser point cloud targets in the monitoring area is more than that of the target locomotive when a pedestrian or a small motor vehicle passes through the crossing, and then the target point characteristics of the locomotive passing through the crossing are compared, and when six or more target points exist in two or more adjacent areas in three continuous scanning periods, the locomotive passing through can be obtained;
Locomotive running direction discrimination and locomotive speed calculation
The method comprises the steps of generating a locomotive position time sequence diagram of a crossing uplink and downlink radar by judging the characteristic of laser point cloud appearing in the radar, obtaining the running direction of the locomotive according to the change of the high and low positions of the time sequence diagram, and calculating the running speed of the locomotive according to the actual distance of a divided area detection range and the time interval of adjacent state change in a time state sequence of the locomotive.
Preferably, the laser radar is also called a two-dimensional laser scanner, the road junction area monitoring works in the laser radar, the two-dimensional laser scanner is used for acquiring depth data of surrounding scenes and moving targets and monitoring configuration data appointed by a user for analysis and processing, and various targets in the scenes are detected, positioned and tracked to realize the monitoring of a plurality of plane areas.
Preferably, the system can detect the behavior of the locomotive based on the locomotive behavior data detected by the laser radar in real time, and acquire the real-time running direction and the current average running speed of the locomotive.
Preferably, in the practical application of the area monitoring function, the area group required to be measured needs to be edited and set according to the actual field scene, so that certain fixed entity targets are prevented from being always located in the monitoring area to trigger the monitoring signal, so that area background cutting is a very important block in the area monitoring function configuration work, and meanwhile, the use effect of area monitoring is greatly influenced.
Preferably, a method for using a railway crossing locomotive detection system based on laser scanning is characterized by comprising the following steps:
Firstly, dividing a railway crossing background area and initializing a background contour, collecting real-time locomotive passing state scanning data of a railway crossing or a crossing at present, accumulating the scanning data of a current real-time scene, and then detecting the approximate background contour of the current crossing scene by using a crossing background self-learning algorithm so as to extract background data of area monitoring, and detecting a crossing object locomotive by using a locomotive area monitoring algorithm;
Extracting all target objects in a monitoring area group in a scene through real-time scanning data, classifying and screening the existing monitoring targets according to the laser point cloud characteristics, eliminating irrelevant targets, and detecting effective targets;
And thirdly, roughly obtaining the running direction of the currently detected locomotive by utilizing a locomotive intrusion detection algorithm according to the data characteristics of the laser point cloud, and roughly detecting the average speed of the locomotive.
Preferably, the crossing background self-learning algorithm essentially utilizes running dynamic data, that is, real-time scanning data of the current railway crossing scene is accumulated for a period of time, so that the background contour of the current scene is detected, and effective background data of regional monitoring is generated according to the background contour.
Preferably, in the locomotive area monitoring algorithm, the monitoring target needs to be segmented, so as to judge the monitoring signal, and the current passing target is operated according to the set laser point cloud parameter, so that the irrelevant target is removed, and the correct monitoring signal is output.
Preferably, the locomotive is large in size and only exists in a rail area, and the locomotive is allowed to pass only under the condition that a road junction is closed, so that the time and the place of occurrence are relatively fixed, the data of the laser point cloud are relatively dense because the locomotive is large in size, a relatively reasonable target area is selected during design, and the data characteristics of the laser point cloud in the detection range are more dense than those of pedestrians or small vehicles when the locomotive object is detected to be close.
Preferably, the time sequence of the locomotive position generated by the laser radar generating the laser point cloud is a characteristic for reflecting the running direction of the locomotive, the time sequence of the position of the ascending and descending radar is changed in high and low order, the condition that the locomotive passes through the radar of the current crossing and the time sequence of the position of the radar of the other side is combined can be known, the running direction of the locomotive can be obtained, and the approximate speed of the locomotive can be obtained according to the actual distance of the divided area detection range and the time interval of the adjacent state change in the time state sequence of the locomotive.
Preferably, the crossing area division refers to dividing an area to be detected into a plurality of areas with equal size, calculating the number of target points in each divided area in the same scanning period, and if six or more target points in two or more adjacent small areas act in three continuous scanning periods, it is possible to determine that a locomotive passes at the target railway crossing, otherwise, when the number of the appearing target points is small, it is possible to predict that a pedestrian or a small locomotive passes.
Compared with the prior art, the invention has the beneficial effects that: the railway crossing locomotive detection system based on laser scanning adopts a laser radar detection mode, is less affected by the environment, has high reliability, is least affected by weather, has more accurate detection results, and is more suitable for the close-range locomotive judgment of the railway crossing.
1. The laser radar detection utilizes emitted laser beams, reflects the laser beams after encountering detected target objects, analyzes laser reflected beam data to analyze the target objects, can actively detect and obtain high-precision laser point cloud information on the surfaces of the objects, and calculates the accurate distance between the detected objects and the radar. The laser radar is not easy to be interfered and has high stability, so that the detection result is more accurate;
2. The railway crossing locomotive detection system based on laser scanning can receive the real-time traffic state scanning data of the railway crossing locomotive from the laser radar in real time, store the data and provide a data base for locomotive behavior analysis and abnormal behavior reminding;
3. The data acquisition laser radar related by the invention is simple and convenient to install, and the automatic recognition of the state of the locomotive with the road entrance behavior can be realized by utilizing a pre-designed detection algorithm.
Drawings
FIG. 1 is a schematic diagram of a railway crossing locomotive detection system based on laser scanning according to the present invention;
fig. 2 is a schematic flow chart of a railway crossing locomotive detection system based on laser scanning.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-2, the present invention provides a technical solution: the railway crossing locomotive detection system based on laser scanning consists of a laser radar and a corresponding locomotive detection algorithm, wherein the working flow is as follows: two laser radars of a railway crossing uplink and downlink area scan the crossing to form laser point cloud data for initializing a crossing background contour, then dividing a detection range of an area where a locomotive enters and leaves the crossing, detecting a crossing locomotive object by adopting a crossing background self-learning algorithm and a locomotive area detection algorithm, identifying and classifying a current scanning object target by the laser point cloud characteristic data acquired by the laser radars, establishing a time state sequence of entering and leaving the crossing of the locomotive according to the scanning data of the crossing detection area of the crossing of the locomotive, and then analyzing the running direction and the approximate speed of the locomotive at the crossing, wherein the detection method comprises the following steps of railway crossing detection data initialization, locomotive object detection, locomotive running direction judgment and locomotive speed calculation, and the specific implementation scheme is as follows:
Embodiment one:
railway crossing detection data initialization
The regional group is used for describing locomotive data in the coverage area of each region, the crossing locomotive region monitoring function is based on regional group work, the shape region needing to be scanned by the laser radar at the railway crossing is subjected to self-defining drawing, and the shape of the region is adjusted and the edges are refined according to the on-site working condition of the railway crossing to obtain the required regional group. The regional division adopts a 'depth range' expression mode, main data of a data structure comprises a start angle theta k,0 and an end angle theta k,n of a regional scanning angle, the depth range comprises a start depth and an end depth, the range represents the range of the ray section distance of the region on the scanning angle, the depth measurement data on the scanning angle only belong to the region when being positioned in the range, after the geometric definition of the regional group is completed, the regions in the group are converted into a uniform internal data structure, polar coordinate expression is adopted, the efficiency of a regional monitoring algorithm is improved, the self-learning of a crossing background and the initialization of a background profile of a locomotive regional monitoring algorithm are further carried out, wherein, the crossing background self-learning algorithm is used for generating the background B of the current scene, B also adopts a 'depth range' expression mode under polar coordinates, the scanning angle range is the original scanning angle range of the scanner, the depth range on each scanning angle theta s is [0, br s],brs is the depth value of the stable background detected on the scanning angle, during the background learning, the br s is normally modeled, namely, the measured value rho of the current moment t is calculated on the scanning angle theta s, the crossing background learning algorithm adopts a standard normal distribution parameter calculation method, using real-time measurement data to iteratively solve br s and sigma 2 s, using 3 sigma s as a threshold value to eliminate background noise data, after finishing iterative solution within a specified learning time, using the obtained br s to establish a background area representation, and for any activated monitoring area group phi, having an operation state omega; locomotive area monitoring algorithm: acquiring all effective target sets t,Оt＝{Objt,l,l},l＝0,…,Lt -1 with L being the effective target number of the current time t in the activated monitoring area group, wherein O t calculates on the total activation area X, X is the union of omega, calculates in the initialization stage of the monitoring algorithm, the target segmentation algorithm firstly acquires the number of measuring points with P t * of all point sets Pt *＝{Pt *,sj'},sj'∈[0,S-1],j＝0,…,Jt-1,Jt in P t * from the measuring data sets t＝{ρt, s of the current time t, then carries out connectivity clustering on P t * according to the scanning angle sequence and the fixed distance threshold epsilon, the obtained data structure representation of the object t、Objt,l includes, in addition to all measurement point sets t,l belonging to the target, an circumscribed rectangle Rect t,l of the measurement point sets and a diagonal length lambda t,l of Rect t,l, which are used as a basis for determining whether the effective target triggers the monitoring signal in the next step.
Locomotive object detection
Since a laser beam emitted by a radar can only correspond to the distance, reflection intensity and the like of a point on the surface of a detected object, and two-dimensional information of the object is required to be scanned, the emitted beam can be scanned in the horizontal direction to obtain two-dimensional laser point cloud data with wide vision and high resolution, the data obtained from the laser radar is polar coordinate data taking the center as the midpoint, and since the obtained radar data is more, a part of measuring noise is contained, and in order to facilitate the detection of the later vehicle intrusion, the coordinate transformation is required to be performed on the measuring result, namely, the polar coordinate is converted into reference rectangular coordinate data, and then the filtering processing is performed on the original measuring data point to remove the influence caused by the measuring noise. The method is characterized in that the crossing passing locomotive is large in size and can only run on rails, namely, the area needing to be monitored is divided into a plurality of equal areas, and the number of target points in each divided area in the same scanning period is calculated, for example, when six or more target points in two or more adjacent areas act in three or more continuous scanning periods, the number of locomotives passing at the crossing of a target railway can be obtained, and conversely, when the number of the target points appearing is small, the number of the locomotives passing at the crossing of the target railway can be judged, and then the response time of the crossing passing at the crossing of the target point exceeds a set threshold value.
Locomotive running direction discrimination and locomotive speed calculation
The detection system composed of double radars is utilized, a time state sequence of the locomotive entering and exiting the road junction is established according to the scanning data of the locomotive entering and exiting the road junction detection area, the instantaneous running speed of the locomotive is calculated according to the actual distance of the divided area detection range and the time interval of adjacent state change in the time state sequence of the locomotive, and the average running speed of the locomotive is obtained after multiple value calculation.
Working principle: according to the architecture shown in fig. 1 and 2, the system implementation comprises the following steps:
S1, installing two laser radars with 270-degree scanning ranges in an uplink area and a downlink area of a railway crossing, scanning the crossing to form laser point cloud data to initialize a crossing background contour, and then dividing an area detection range of a locomotive entering and exiting the crossing;
s2, after locomotive detection initialization is completed, real-time scanning is conducted on a railway crossing to form dynamic data of a crossing area, and a crossing background self-learning algorithm and a locomotive area detection algorithm are adopted to detect a crossing locomotive object;
S3, extracting all targets in the monitoring area group in the scene from the real-time scanning data, and identifying and classifying the current scanning object targets according to the concentration and the number of the laser point cloud characteristic data acquired by the laser radar;
S4, according to the scanning data of the detection area of the entrance of the locomotive and the detection area of the exit of the locomotive, establishing a time state sequence of the entrance of the locomotive and the exit of the locomotive, analyzing the running direction of the locomotive at the entrance, and finally calculating the running speed of the locomotive according to the actual distance of the detection range of the divided area and the time interval of the change of the adjacent states in the time state sequence of the locomotive;
aiming at the current situations that the working condition environment of a railway crossing is complex, the mutual passing risk of pedestrians and locomotives is high, and the real-time monitoring of complex traffic roads is difficult, the invention provides a method capable of monitoring the running behavior and speed detection of the locomotive at the railway crossing based on the deep fusion technology of the area monitoring function (a crossing background self-learning algorithm and a locomotive area monitoring algorithm) and the locomotive behavior detection algorithm, and the method is combined to design a railway crossing locomotive detection system based on laser scanning.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. A railway crossing locomotive detection system based on laser scanning comprises a laser radar and a locomotive detection algorithm, wherein the locomotive detection algorithm comprises railway crossing detection data initialization, locomotive object detection and locomotive running direction judging mechanical locomotive speed calculation, and the system comprises the following components:
railway crossing detection data initialization
The method comprises the steps of carrying out custom drawing on a road junction laser radar scanning area, then carrying out adjustment and edge finishing on the shape of the area according to the on-site working condition of a railway road junction to obtain a required area outline, carrying out area division by adopting a 'depth range' expression mode, wherein main data of a data structure comprise a start angle and a stop angle of an area scanning angle, the depth range comprises a start depth and a stop depth, the range represents a ray range distance range of the area on the scanning angle, depth measurement data on the scanning angle only belong to the area when the depth measurement data are positioned in the range, after the geometric definition of an area group is completed, the area in the group is converted into a uniform internal data structure, adopting polar coordinate representation to improve the efficiency of an area monitoring algorithm, and further initializing the background outline by utilizing a road junction background self-learning algorithm and a locomotive area monitoring algorithm, wherein the road junction background self-learning algorithm refers to the accumulation of real-time scanning data of a current scene for a period of time so as to detect the background outline of the current scene, and accordingly generating background data of area monitoring; the locomotive area monitoring algorithm is used for extracting all targets in a monitoring area group in a scene from real-time scanning data, classifying and identifying current monitoring targets according to operation control parameters, eliminating irrelevant targets and detecting effective monitoring targets;
In the practical application of the area monitoring function, the area group required to be measured is required to be edited and set according to the actual scene, so that certain fixed entity targets are prevented from being always positioned in the monitoring area to trigger the monitoring signal, therefore, area background cutting is a quite important block in the area monitoring function configuration work, and meanwhile, the use effect of area monitoring is greatly influenced;
Locomotive object detection
In order to scan the two-dimensional information of a real object, the laser radar needs to scan and detect the emitted laser beam in the horizontal direction to obtain laser point cloud data which has huge data, wide viewing area, high resolution and certain measurement noise characteristics, because the locomotive has larger volume factors, the laser point cloud data obtained by the radar can be more than other detection targets and data points can be more dense, aiming at a railway crossing area needing to be detected, the laser point cloud characteristics when a target locomotive passes are observed, the number of the laser point cloud targets in a monitoring area is more than the number of the laser point cloud targets when a pedestrian or a small motor vehicle passes through the crossing when the locomotive approaches, further the target point characteristics of the locomotive passing through the crossing are compared, and when six or more target points exist in two or more adjacent areas in three continuous scanning periods, the crossing can be obtained;
Locomotive running direction discrimination and locomotive speed calculation
The method comprises the steps of generating a locomotive position time sequence diagram of a crossing uplink and downlink radar by judging the characteristic of laser point cloud appearing in the radar, obtaining the running direction of the locomotive according to the change of the high and low positions of the time sequence diagram, and calculating the running speed of the locomotive according to the actual distance of a divided area detection range and the time interval of adjacent state change in a time state sequence of the locomotive.
2. The railroad grade crossing locomotive detection system based on laser scanning of claim 1, wherein: the laser radar is also called a two-dimensional laser scanner, the road junction area monitoring works in the laser radar, the two-dimensional laser scanner is used for acquiring depth data of surrounding scenes and moving targets and monitoring configuration data appointed by a user for analysis and processing, and various targets in the scenes are detected, positioned and tracked, so that the monitoring of a plurality of plane areas is realized.
3. The railroad grade crossing locomotive detection system based on laser scanning of claim 1, wherein: the system can detect the behavior of the locomotive based on the locomotive behavior data detected by the laser radar in real time, and acquire the real-time running direction and the current average running speed of the locomotive.
4. The method of using a railroad grade crossing locomotive inspection system based on laser scanning of claim 1, comprising the steps of:
firstly, dividing a railway crossing background area and initializing a background contour, collecting real-time locomotive passing state scanning data of a railway crossing or a crossing, accumulating a section of scanning data of a current real-time scene, and then detecting the background contour of the current crossing scene by using a crossing background self-learning algorithm so as to extract background data of area monitoring, and detecting a crossing object locomotive by using a locomotive area monitoring algorithm;
Extracting all target objects in a monitoring area group in a scene through real-time scanning data, classifying and screening the existing monitoring targets according to the laser point cloud characteristics, eliminating irrelevant targets, and detecting effective targets;
step three, according to the data characteristics of the laser point cloud, the running direction of the currently detected locomotive is obtained by utilizing a locomotive intrusion detection algorithm, and meanwhile, the average speed of the locomotive is detected;
The locomotive is large in size and only exists in a rail area, and the locomotive is allowed to pass only under the condition that a road junction is closed, so that the time and the place of occurrence are relatively fixed, the data of the laser point cloud are relatively dense because the locomotive is large in size, a relatively reasonable target area is selected during design, and when a locomotive object is detected to be close, the data characteristics of the laser point cloud in the detection range are more dense than those of pedestrians or small automobiles during passing.
5. The method for using the railway crossing locomotive detection system based on laser scanning as claimed in claim 4, wherein the method comprises the following steps: the crossing background self-learning algorithm essentially utilizes running dynamic data, namely, real-time scanning data of the current railway crossing scene is accumulated for a period of time, so that the background contour of the current scene is detected, and effective background data for regional monitoring is generated according to the background contour.
6. The method for using the railway crossing locomotive detection system based on laser scanning as claimed in claim 4, wherein the method comprises the following steps: in the locomotive area monitoring algorithm, a monitoring target is required to be segmented, so that a monitoring signal is judged, a current passing target is operated according to a set laser point cloud parameter, an irrelevant target is removed, and a correct monitoring signal is output.
7. The method for using the railway crossing locomotive detection system based on laser scanning as claimed in claim 4, wherein the method comprises the following steps: the time sequence of the locomotive position generated by the laser radar generating the laser point cloud is a characteristic for reflecting the running direction of the locomotive, the time sequence of the position of the ascending and descending radar is changed in high and low order, the condition that the locomotive passes through the radar of the current crossing and the time sequence of the position of the radar of the other side is combined can be known, the running direction of the locomotive can be obtained, and the speed of the locomotive can be obtained according to the actual distance of the divided area detection range and the time interval of the adjacent state change in the time state sequence of the locomotive.
8. The method for using the railway crossing locomotive detection system based on laser scanning as claimed in claim 4, wherein the method comprises the following steps: the railway crossing background area division refers to dividing an area to be detected into a plurality of areas with the same size, calculating the number of target points in each divided area in the same scanning period, and judging that a locomotive passes at a target railway crossing when six or more target points in two or more adjacent small areas act in three continuous scanning periods, otherwise, estimating that a pedestrian or a small locomotive passes when the number of the target points is less.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110354002.0A CN113514848B (en) | 2021-04-01 | 2021-04-01 | Railway crossing locomotive detecting system based on laser scanning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110354002.0A CN113514848B (en) | 2021-04-01 | 2021-04-01 | Railway crossing locomotive detecting system based on laser scanning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113514848A CN113514848A (en) | 2021-10-19 |
| CN113514848B true CN113514848B (en) | 2024-05-31 |
Family
ID=78061284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110354002.0A Active CN113514848B (en) | 2021-04-01 | 2021-04-01 | Railway crossing locomotive detecting system based on laser scanning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113514848B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114322828A (en) * | 2021-12-15 | 2022-04-12 | 云南昆钢电子信息科技有限公司 | Point cloud sampling system and method based on laser radar scanning angle switching |
| CN114638853B (en) * | 2022-05-18 | 2022-09-02 | 广东工业大学 | Method and system for filtering background based on roadside solid-state laser radar |
| CN114782469B (en) * | 2022-06-16 | 2022-08-19 | 西南交通大学 | Method and device for identifying degree of congestion of public traffic, electronic device and storage medium |
| CN117092658B (en) * | 2023-10-20 | 2023-12-26 | 徐州瑞控机电科技有限公司 | Target intelligent detection positioning method based on laser radar |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005212553A (en) * | 2004-01-28 | 2005-08-11 | East Japan Railway Co | Detecting device and detecting method |
| JP2010042724A (en) * | 2008-08-11 | 2010-02-25 | Ihi Corp | Device and method for detecting railroad crossing obstruction |
| CN108594247A (en) * | 2018-03-05 | 2018-09-28 | 湖北三江航天红峰控制有限公司 | A kind of railway contact net measuring method and apparatus based on bidifly optical scanning |
| CN109229161A (en) * | 2018-08-27 | 2019-01-18 | 杭州登元科技有限公司 | Intelligent machine vehicle detection control apparatus |
| CN110161524A (en) * | 2019-05-05 | 2019-08-23 | 中国科学院半导体研究所 | Prevent train from passing through the scanning radar judgment method of wrong report |
| CN110466562A (en) * | 2019-08-14 | 2019-11-19 | 南京慧尔视智能科技有限公司 | There are detection devices and method for a kind of level crossing based on infrared laser and microwave |
| CN110481601A (en) * | 2019-09-04 | 2019-11-22 | 深圳市镭神智能系统有限公司 | A kind of rail detection system |
| KR102188567B1 (en) * | 2020-05-18 | 2020-12-10 | (주) 안세기술 | System for monitoring the road using 3 dimension laser scanner |
-
2021
- 2021-04-01 CN CN202110354002.0A patent/CN113514848B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005212553A (en) * | 2004-01-28 | 2005-08-11 | East Japan Railway Co | Detecting device and detecting method |
| JP2010042724A (en) * | 2008-08-11 | 2010-02-25 | Ihi Corp | Device and method for detecting railroad crossing obstruction |
| CN108594247A (en) * | 2018-03-05 | 2018-09-28 | 湖北三江航天红峰控制有限公司 | A kind of railway contact net measuring method and apparatus based on bidifly optical scanning |
| CN109229161A (en) * | 2018-08-27 | 2019-01-18 | 杭州登元科技有限公司 | Intelligent machine vehicle detection control apparatus |
| CN110161524A (en) * | 2019-05-05 | 2019-08-23 | 中国科学院半导体研究所 | Prevent train from passing through the scanning radar judgment method of wrong report |
| CN110466562A (en) * | 2019-08-14 | 2019-11-19 | 南京慧尔视智能科技有限公司 | There are detection devices and method for a kind of level crossing based on infrared laser and microwave |
| CN110481601A (en) * | 2019-09-04 | 2019-11-22 | 深圳市镭神智能系统有限公司 | A kind of rail detection system |
| KR102188567B1 (en) * | 2020-05-18 | 2020-12-10 | (주) 안세기술 | System for monitoring the road using 3 dimension laser scanner |
Non-Patent Citations (3)
| Title |
|---|
| A NEW RAILYARD SAFETY APPROACH FOR DETECTION AND TRACKING OF PERSONNEL AND DYNAMIC OBJECTS USING SOFTWARE-DEFINED RADAR;Banerjee, S等;Proceedings of the ASME Joint Rail Conference;第1-10页 * |
| 基于激光测量技术的铁轨障碍物检测系统研究;蔚增岩;中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑(第01期);B026-203页 * |
| 基于激光雷达的集装箱堆场智能安全防护与辅助作业系统;徐家等;港口科技·中国港口协会科学技术奖优秀成果(第06期);第1-5、27页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113514848A (en) | 2021-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113514848B (en) | Railway crossing locomotive detecting system based on laser scanning | |
| US10451720B2 (en) | Object detection apparatus and object detection method | |
| CN109657698B (en) | Magnetic suspension track obstacle detection method based on point cloud | |
| Dominguez et al. | LIDAR based perception solution for autonomous vehicles | |
| CN103714363B (en) | A kind of motor vehicle exhaust smoke video identification system | |
| KR100459476B1 (en) | Apparatus and method for queue length of vehicle to measure | |
| CN106127113A (en) | A kind of road track line detecting method based on three-dimensional laser radar | |
| CN113345237A (en) | Lane-changing identification and prediction method, system, equipment and storage medium for extracting vehicle track by using roadside laser radar data | |
| CN105404844A (en) | Road boundary detection method based on multi-line laser radar | |
| CN104778444A (en) | Method for analyzing apparent characteristic of vehicle image in road scene | |
| CN109444872B (en) | Pedestrian and vehicle distinguishing method and device, computer equipment and storage medium | |
| CN114003849B (en) | Multi-lane non-contact type automatic calculation method and system for vehicle axle number | |
| CN114415171A (en) | Automobile travelable area detection method based on 4D millimeter wave radar | |
| CN114814826B (en) | Radar orbit area environment sensing method based on target grid | |
| CN114155720B (en) | Vehicle detection and track prediction method for roadside laser radar | |
| CN116699602A (en) | Target detection system and method based on millimeter wave radar and camera fusion | |
| CN110554409B (en) | Concave obstacle detection method and system | |
| KR20230101560A (en) | Vehicle lidar system and object detecting method thereof | |
| CN114842166A (en) | Negative obstacle detection method, system, medium, and apparatus applied to structured road | |
| Chen et al. | A Novel Background Filtering Method with Automatic Parameter Adjustment for Real-Time Roadside LiDAR Sensing System | |
| CN113654632A (en) | Railway freight transportation geometric overrun early warning system based on high-speed laser radar | |
| Huang et al. | An efficient multi-threshold selection method for lane detection based on lidar | |
| CN115546595A (en) | Track tracking method and system based on fusion sensing of laser radar and camera | |
| Sun et al. | Background Extraction and Objects Segmentation with 3D Roadside LiDAR under Snowy Weather | |
| CN115116034A (en) | Method, device and system for detecting pedestrians at night |
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 |