CN109164443A - Rail track foreign matter detecting method and system based on radar and image analysis - Google Patents
Rail track foreign matter detecting method and system based on radar and image analysis Download PDFInfo
- Publication number
- CN109164443A CN109164443A CN201810978155.0A CN201810978155A CN109164443A CN 109164443 A CN109164443 A CN 109164443A CN 201810978155 A CN201810978155 A CN 201810978155A CN 109164443 A CN109164443 A CN 109164443A
- Authority
- CN
- China
- Prior art keywords
- target
- foreign matter
- rail track
- image
- mobile
- 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
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
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
- G01S13/867—Combination of radar systems with cameras
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a kind of rail track foreign matter detecting method analyzed based on radar and high-definition image and systems, by the advantage for making full use of microwave radar detecting real-time moving target, reuse the real-time high-definition video stream that high-definition video equipment obtains moving target, moving target is analyzed then in conjunction with the two, judges whether it is the foreign matter target for needing to carry out early warning.And the early warning of different stage can be carried out, for size, the type of different foreign matter targets to improve detection accuracy and response speed to invasion foreign matter.Compared with prior art, the present invention, which can be realized, carries out timely, accurately detecting and early warning to the intrusion foreign matter in rail track, omission factor, rate of false alarm are effectively reduced, avoids due to foreign body intrusion and major accident occurs, is had a good application prospect in rail track security fields.
Description
Technical field
The invention belongs to rail traffic security technology areas, are related to a kind of rail track based on radar and image analysis
Foreign matter detecting method and system.
Background technique
In recent years, as railway construction in China dynamics increasingly increases, especially in terms of high speed railway construction, country's investment
A large amount of manpower, financial resources carry out development & construction.But also the high-speed rail technology in China occupy world forefront, become one of China
New business card.It is also higher and higher to the safety requirements of rail track meanwhile with the continuous improvement of train speed.And ensure train
Operational safety other than the reliable in quality for requiring train and railway bed itself, the resident of Along Railway, animal and it is various because
The foreign matters such as branch, falling rocks caused by bad weather or the rubbish artificially abandoned invade railway clearance or even various artificial malice
Damage, has seriously affected the development of railway operation safety and railway transport economy.Therefore, railway foreign body intrusion detection problem becomes
The hot issue of China Railway due care.
Traditional rail track foreign bodies detection mainly relies on manpower, carries out continual line by largely patrolling passerby person
Road inspection whether there is foreign matter to check on rail track.Although this mode can accurately detect very much on route whether
There are foreign matters, but need a large amount of manpower, and since the travel speed of people is limited, and foreign body intrusion is with sudden, random
The features such as restraining, is unpredictable, so that the discovery of foreign matter is not prompt enough, the cleaning eventually led to foreign matter is not rapid enough.And
With the rapid development of high-definition video monitoring technology, the foreign body intrusion detection technique based on video monitoring has been obtained more and more
Using.But high-definition video equipment then there may be blind area, be easy by the such environmental effects such as weather illumination, can not be accurately
Obtain the accurate location of invasion foreign matter, the disadvantages of detection algorithm is complicated so as to the detection of foreign body intrusion not enough precisely and
Reliably, there are serious alert locations inaccuracy, wrong report rate of failing to report is higher, alarm not in time the problems such as.
To solve the above problems, Through Several Survey Measure has been used to carry out in certain rail track foreign matter detection systems simultaneously
Deployment is to avoid the above problem, such as using double power grid, inclination angle sensing, infrared emission, the radar exploration techniques, laser curtain wall, high definition
Two or more in the various technological means such as video are detected to cooperate, to improve the precision and reliability of detection.
Such as with the development of commerical radar technology, to some new detection technique means occur, such as laser radar skill
Art and microwave radar technology.Wherein laser infrared radar imaging technology have blind area it is small, high-precision, high-resolution advantage, utilize it
Image-forming information carries out target identification, also can be used to carry out foreign matter identification on rail track to a certain degree, but laser radar is same
Influence of the sample vulnerable to weather conditions such as ice and snow.The rail track that can not adapt to China's complexity uses bad border.And microwave radar is
A kind of to detect the electronic equipment of target to send electromagnetic wave and receive the reflection echo of object, microwave radar passes through detection microwave
The phase change of signal is able to achieve the speed to target mobile in monitoring region, the function that position coordinate is detected.
If CN 107697102A discloses information collection, identification and the warning system of a kind of high-speed railway foreign body intrusion,
The identification of foreign matter is carried out using radar and Image Acquisition simultaneously, basic ideas are using different objects to the anti-of radar electromagnetic wave
The long different mode of ejected wave whether there is foreign matter to determine on railway track, the image of foreign matter is then obtained using area array CCD,
And the size of foreign matter is calculated according to accounting size of the foreign matter in the battle array of the face CCD.Although this mode under certain conditions can be with
It detects with the presence or absence of foreign matter on railway track, but its defect is also apparent, (1) same object is under different weather to electromagnetism
The wavelength reflection situation of wave is different, is such as directed to for railway track, if heavy rain, even heavy snow weather rail track are by ice
Freeze, is necessarily affected to the wavelength of reflection of electromagnetic wave, is highly prone to the influence of weather in this way.(2) it is only
The size of foreign matter is obtained using area array CCD, so that the details profile information of foreign matter can not be got, foreign matter can not be carried out quasi-
True classification.In addition, the difference of distance also will affect imaging size of the object on CCD, thus only with imaging size
The actual size of object can not accurately be obtained.(3) its equipment installation position requires to be lower than track, so as to cause the work of radar equipment
It with being limited in scope, and is easy by artificial destruction, and actual foreign matter identification, warning effect can not be played.
Summary of the invention
Goal of the invention: in view of the deficiencies of the prior art, it is an object of that present invention to provide one kind to be based on radar and image analysis
Rail track foreign matter detecting method and system, it is contemplated that all there is a dynamic phagocytic process in any foreign body intrusion, with
And the characteristic that microwave radar is extremely sensitive to moving object, the phagocytic process of foreign matter is detected by microwave radar, and detecting
After foreign matter, cooperation high clear video image carries out the analysis of the type, size of foreign matter, to realize the danger coefficient to foreign matter
It is calculated, and issues the pre-warning signal of different stage accordingly, alarmed to realize foreign matter all on rail track.
Technical solution: for achieving the above object, the present invention adopts the following technical scheme:
A kind of rail track foreign matter detecting method based on radar and image analysis, includes the following steps:
(1) by being deployed in the microwave radar equipment in rail track roadside, whether there is shifting in its monitoring range of real-time detection
Moving-target, and send the co-ordinate position information of all mobile targets in corresponding high-definition tripod video camera；
(2) it is opposite to calculate mobile target according to the relative position of itself and microwave radar equipment for high-definition tripod video camera
In the position of high-definition camera, shooting direction and focal length are adjusted, mobile target image (video flowing) is continuously shot, until target is stopped
Only move or have moved out the reconnaissance range of microwave radar equipment；
(3) foreign bodies detection is carried out according to the realtime graphic (video flowing) of high-definition tripod video camera shooting, comprising:
The position moved target on the image and chamfered shape are obtained using the image shot every time；
The full-size(d) of mobile target is calculated according to video camera imaging principle；
It is carried out curve fitting according to the position coordinates of mobile target and obtains the space motion path equation of mobile target；
The type of target is determined in conjunction with the termination condition that the space motion path equation and picture of mobile target are continuously shot,
If monitoring range or target landing position that mobile target has moved out radar equipment outside the monitoring range of radar equipment,
Think that target does not have risk, excludes target；If target Continuous moves in microwave radar monitoring range or landing position exists
In monitoring range, then images match analysis is carried out according to the profile of target and texture, obtain foreign matter type and calculate foreign matter target
Danger coefficient；
(4) Warning Service device is sent by foreign matter parameter, danger coefficient and realtime graphic, Warning Service device is according to predefined
Early warning rule, generate and send early warning order to corresponding prior-warning device or the platform of docking.
Further, in the step (3), by carrying out gray processing to two images being continuously shot every time and making poor fortune
It calculates, then carries out morphologic expansion process, obtain the position moved target on the image and chamfered shape.
Further, in the step (3), the full-size(d) calculation formula of mobile target are as follows:
Wherein, W is the true cross-sectional width of mobile target, and H is the true depth of section of mobile target, and w is mobile target
Width on the image, h are the height of mobile target on the image, and D is shooting time shift moving-target at a distance from video camera, and L is
The image distance of mobile target.
Further, in the step (3), for simultaneously have multiple targets by microwave radar equipment sensing to the case where,
Judge whether it is the same target according to position coordinates and the movement velocity that the target last time is detected is moved, specifically
Judgment formula are as follows:
Wherein, P0It is to be detected position, P the last time for mobile target1It is present position, V for mobile target0For
Movement velocity when the mobile target last time is detected, T1、T0Specific moment when being detected for mobile target.
Further, in the step (3), described image the matching analysis uses the image classification side based on deep learning
Method, all angles image by collecting various possible invasion foreign matters are added deep learning model, carry out model training, divided
Class model；The contour images input disaggregated model of the mobile target taken is obtained into foreign matter type.
Further, in the step (3), the calculation formula of the danger coefficient R of foreign matter target are as follows:
Wherein, R0Danger coefficient corresponding to foreign matter type for configured in advance, S are the sectional area size of foreign matter, and D is different
Object size at a distance from rail track rail center.
A kind of rail track foreign matter detection system based on radar and image analysis that another aspect of the present invention provides, packet
It includes: being mounted on the microwave radar equipment in rail track roadside, it is corresponding with microwave radar equipment to be mounted on rail track roadside
High-definition tripod video camera and Warning Service device；
The microwave radar equipment whether there is mobile target for detecting in its reconnaissance range, and mobile target
Speed, position three-dimensional coordinate information, and will send information to corresponding high-definition tripod video camera；
The high-definition tripod video camera, for receive microwave radar equipment transmission moving target position after, according to
The relative position of itself and microwave radar equipment calculates position of the mobile target relative to high-definition camera, adjusts shooting side
To and focal length, mobile target image is continuously shot, until target stops detecting that is mobile or having moved out microwave radar equipment
Range；And foreign bodies detection is carried out by realtime graphic of the built-in foreign bodies detection module to shooting, lead to after detecting foreign matter
It crosses network and foreign matter warning information is sent to Warning Service device, wherein foreign matter detecting method includes: to be obtained using the image of shooting
Mobile target position on the image and chamfered shape；The full-size(d) of mobile target is calculated according to video camera imaging principle；Root
It carries out curve fitting according to the position coordinates of mobile target and obtains the space motion path equation of mobile target；In conjunction with mobile target
The termination condition that space motion path equation and picture are continuously shot determines the type of target, if mobile target has moved out thunder
Up to equipment monitoring range or target landing position outside the monitoring range of radar equipment, then it is assumed that target does not have risk,
Exclude target；If target Continuous moves in microwave radar monitoring range or landing position is in monitoring range, according to target
Profile and texture carry out images match analysis, obtain foreign matter type and calculate the danger coefficient of foreign matter target；
The Warning Service device generates early warning order and sends for receiving foreign matter warning information, and according to early warning rule
To the third-party platform for the prior-warning device or docking for being mounted on rail track roadside.
Further, the high-definition tripod video camera, microwave radar equipment and prior-warning device are fixed on peace by rail track
In the upright bar of dress, it is also equipped in the upright bar and is provided for the high-definition tripod video camera, microwave radar equipment and prior-warning device
The peripheral equipment of power supply and network communication.
It further, include high-definition tripod video camera, microwave by being installed a set of by rail track every set distance L
The equipment of radar equipment and prior-warning device completely monitors a rail track, grows wherein distance L is monitored according to microwave radar maximum
The maximum sighting distance of degree and high-definition camera determines.
Further, the reconnaissance range length of the microwave radar equipment be configured to microwave radar equipment adjacent thereto away from
From width configuration is the distance between rail track guardrails of both sides.
The utility model has the advantages that compared with prior art, the present invention has the advantage that
1, the present invention is based on any foreign body intrusions all there is the considerations of dynamic phagocytic process, so as to pass through microwave
All mobile targets in radar equipment real-time detection rail track and its neighboring area, and according to the mobile target detected
Corresponding high-definition camera is gone to the position to obtain the real-time high definition image of mobile target by position, finally combines radar detection
Data and high-definition image analyze mobile target, confirm whether it is true foreign matter target.
2, the present invention saves a large amount of man power and material relative to traditional artificial visiting method of rail track foreign matter.
And for the object detection method for only relying upon video image or laser radar, the present invention using microwave radar equipment come
The position coordinates and movement speed for detecting mobile target, due to the ranging of microwave radar, the performance that tests the speed have it is round-the-clock, full-time
Section, not by dust and smog and it is various it is boisterous influence, be not illuminated by the light the influence of condition yet, therefore, substantially increase different
The precision and reliability of analyte detection.
3, the rail track foreign matter target real-time detection method that the present invention realizes, greatly improves to foreign matter in rail track
The detection accuracy and real-time of target.Reduce omission factor, rate of false alarm.And related personnel can be carried out in a manner of significant effective
Early warning, therefore have a good application prospect in rail track security fields.
Detailed description of the invention
Fig. 1 is the method flow diagram of the embodiment of the present invention.
Fig. 2 is the system framework figure of the embodiment of the present invention.
Fig. 3 is that the device context in the embodiment of the present invention disposes schematic diagram.
Fig. 4 is microwave radar and high-definition tripod video camera collaborative work schematic diagram in the embodiment of the present invention.
Fig. 5 is parabolic target invasion procedure schematic diagram in the embodiment of the present invention.
Fig. 6 is the detail flowchart of the foreign matter examinations method of the embodiment of the present invention.
Specific embodiment
Combined with specific embodiments below and its technical solution of the present invention is clearly and completely described in attached drawing.It can be with
Understand, it is not whole that shown accompanying drawings and embodiments, which are a part of the invention, does not limit implementation field of the invention yet
Scape and range.
As shown in Figure 1, a kind of rail track foreign bodies detection based on radar and image analysis disclosed by the embodiments of the present invention
Method mainly includes the following steps:
(1) by being deployed in the microwave radar equipment in rail track roadside, whether there is shifting in its monitoring range of real-time detection
Moving-target, and send the co-ordinate position information of all mobile targets in corresponding high-definition tripod video camera；
(2) it is opposite to calculate mobile target according to the relative position of itself and microwave radar equipment for high-definition tripod video camera
In the position of high-definition camera, its video pictures center is moved to the position where mobile target by rotary platform, and according to height
The distance of clear video camera to mobile target carries out zoom, is continuously shot the clear detail pictures of mobile target, until target stops
Reconnaissance range that is mobile or having moved out microwave radar equipment；Wherein, high-definition tripod video camera shoots two images every time；
(3) realtime graphic shot according to high-definition tripod video camera carries out mobile mesh in conjunction with the position of mobile target
Target profile, Texture Matching analysis, exclude wrong report target, true foreign matter, confirmation foreign matter type, size etc. are detected, to true
Foreign matter target calculates its danger coefficient according to parameters such as its type, material, sizes；
(4) Warning Service device is sent by foreign matter parameter, danger coefficient and realtime graphic, Warning Service device displays and store
Corresponding foreign matter parameter information, realtime graphic, and according to predefined early warning rule generate and send early warning order to corresponding
Prior-warning device or the third-party platform of docking.
As shown in Fig. 2, a kind of rail track foreign matter based on radar and image analysis disclosed in another embodiment of the present invention
Detection system, comprising: be mounted on the microwave radar equipment in rail track roadside, high-definition tripod corresponding with microwave radar equipment
Video camera, prior-warning device and Warning Service device；
Wherein, microwave radar equipment continuously and uninterruptedly sends microwave radar to its effective reconnaissance range at regular intervals
Signal, and receive the reflection echo of object.It whether there is mobile target in its reconnaissance range for obtaining, and mobile target
Speed, position three-dimensional coordinate information, and send all information in corresponding high-definition tripod video camera.It is wherein micro-
The length and width of effective reconnaissance range of wave radar can be set according to the actual situation on site, to can avoid due to thunder
Wrong report target is introduced up to reconnaissance range mismatch or leads to foreign matter target missing inspection.
High-definition tripod video camera, can real-time Transmission its real time video image within the vision, by high-definition camera, height
Fast horizontal stage electric machine, foreign bodies detection module composition.Wherein high-definition camera is fixed on high-speed holder, and high-speed holder motor has
Vertical 90 degree, horizontal 360-degree high-speed rotating ability, thus high-definition camera can be adjusted by adjusting the position of high-speed holder
Visual field position.In addition, high-definition camera has the function of zoom zoom, therefore prison can be zoomed in or out according to actual needs
The visual field is controlled, to can get the clear detail pictures of mobile target.High-definition tripod video camera is also equipped with light filling device, therefore, no
Pipe daytime, night can clearly shoot mobile target.The then built-in foreign matter detecting method of foreign bodies detection module, it is first
First calculate speed, the location coordinate information of the mobile target detected from radar equipment.Then, high-speed holder is rotated to target
Position, and by the high clear video images of high-definition camera photographic subjects, finally, by the high-definition image of analyzing target come
Confirm whether the mobile target detected is the foreign matter target for needing early warning, then sends the pre- police uniform of warning information to rear end in this way
Business device.
Warning Service device generates early warning order and is sent to installation for receiving foreign matter warning information, and according to early warning rule
Prior-warning device in rail track roadside, to notify related personnel that countermeasure is taken to avoid that accident occurs.
Prior-warning device, such as dedicated early warning warning lamp, for receiving the early warning order of Warning Service device sending, and with the side of flash of light
Formula carries out early warning to train and railroad maintenance personnel.
Above-mentioned high-definition tripod video camera, microwave radar equipment, alarm warning lamp can by installing upright bar by rail track,
Then equipment is fixed in upright bar according to the best effort height of relevant device, equipment installation and deployment very simple is convenient.And
Each upright bar includes outside all devices installed thereon, such as three kinds of above-mentioned equipment and its corresponding power supply, network router
Peripheral equipment may be defined as a set of equipment.By the way that L installs above-mentioned a set of equipment at a certain distance by rail track, and connect
Continuous uninterrupted installation then can completely monitor a rail track.Wherein distance L can monitor length and height according to microwave radar maximum
The maximum sighting distance of clear video camera determines.
In addition, general ball-shaped cloud platform video camera on the market had both can be used at present in above-mentioned high-definition tripod video camera, it can also
Cooperated using common video camera and high-speed holder equipment, it is not limited to specific equipment form and appearance.
Below with reference to concrete application scene as shown in Figure 3 the present invention will be described in detail plan content.The embodiment of the present invention is public
The rail track foreign matter detection system based on radar and image analysis opened, main includes the microwave for being deployed in rail track trackside
The prior-warning devices such as radar equipment (R1~R#), high-definition tripod video camera (C1~C#), early warning warning lamp and be placed on rear end monitoring in
The Warning Service device of the heart.All of above equipment passes through fiber optic network and is connected to the network.Wherein microwave radar equipment and high definition
PTZ camera, early warning warning lamp can correspond to installation, i.e., in rail track roadside, one microwave radar of L installation at a certain distance is set
A standby, high-definition tripod video camera, an early warning warning lamp.Wherein distance L can be according to the maximum detecting distance and height of radar equipment
It is 1 kilometer as L can be determined in the present embodiment depending on the maximum sighting distance of clear video camera.It can be monitored by continuous uninterrupted installation whole
Rail track, as shown in Figure 3.In addition, microwave radar equipment can configure the length and width of its monitoring range.Therefore, when micro-
When wave radar equipment is continuously installed, the monitoring range length of radar equipment can be configured to it and another radar of the front is set
Standby distance, so that seamless connection, does not also have an impact between each other.And the width of radar equipment monitoring range can be configured to iron
The distance between route road guardrails of both sides can avoid detecting the target other than rail track in this way, to reduce jamming target.
Above-mentioned microwave radar equipment is spaced at regular intervals, if the interval time of the present embodiment is 10 milliseconds, Xiang Qi
Reconnaissance range sends radar microwave signal, and analyzes to obtain all mobile mesh in its reconnaissance range based on microwave echoes signal
Mark.As shown in figure 4, the coordinate of mobile target and speed data are led to rapidly when radar detection equipment R has found mobile target Y
Network is crossed to be sent on corresponding high-definition tripod video camera C.Then, high-definition tripod video camera C quickly changes its monitoring model
It encloses, mobile target Y is placed in its central region, and carry out zoom, to shoot the high clear video image of mobile target Y.And it utilizes
Its built-in foreign bodies detection module carries out image analysis to mobile target Y, with confirm its be normal object (such as the train passed through),
The animal passed through, the foreign matter for still needing to carry out early warning.If needing to carry out the foreign matter of early warning, then according to the position where foreign matter
The information such as coordinate information, classification, size generate warning information and are sent on the Warning Service device of monitoring center, in advance
Alert server then sends early warning order to early warning warning lamp according to corresponding warning level, finally lights warning lamp and carries out early warning, and will
Warning information is sent to rail track maintenance personnel, them is notified to remove foreign matter.Equally, if event is urgent, will such as there be train
By and not enough the time carry out foreign matter removing when, alarm command can be sent on train, them is notified to take measures to carry out hedging.
Secondly, Warning Service device includes early warning control platform and data memory module.Early warning control platform can be according to preparatory
The early warning rule of definition generates pre-warning signal and sends early warning order to early warning warning lamp, while can also dock third-party platform, from
And realize Function Extension, such as the function that short message issues, broadcast and connection railway control center are more advanced.Data store mould
Block, then corresponding real-time video when being used to all warning information, the foreign matter object parameter data and foreign body intrusion that detect
Store in server, so as to the later period examination and it is for statistical analysis etc..
Above-mentioned pre-warning signal can be by foreign matter need according to information such as the location coordinate information of foreign matter, classification, sizes
Emergency level to be processed delimited generally to warn, removing in limited time, the immediately warning levels such as exclusion.If the position of foreign matter is in iron
Road guardrail nearby then may be defined as general alert level, and foreign matter then can define in guardrail between rail and not influencing train operation
To remove rank in limited time, and if foreign matter and influences whether that train operating safety defines on rail or between two tracks
To exclude rank immediately, needs to exclude foreign matter immediately, avoid that accident therefore occurs.
In the embodiment of the present invention, high-definition tripod video camera analyzes mould by high-definition camera, high-speed holder motor, foreign bodies detection
Block composition, wherein foreign bodies detection analysis module is mainly used for the mobile target detected according to radar equipment, takes the photograph in conjunction with high definition
The real-time high definition image of camera come analyze mobile target whether be need the foreign matter target of early warning, and finally by foreign matter target according to
Above-mentioned rule is classified.
As shown in figure 4, the radar equipment R and high-definition tripod video camera C that are mounted in monitoring area, detect radar equipment
Any mobile target measured, such as the target Y in Fig. 4, radar equipment can get its relative to radar installation site coordinate to
AmountAnd high-definition tripod video camera is sent by the vectorial coordinate value, then, high-definition tripod video camera is then according to above-mentioned position
Vector and itself position vector relative to radar equipmentCalculate seat of the mobile target relative to high-definition tripod video camera
Mark vectorAnd the angle that high-speed holder is rotated is calculated according to the coordinate value of the vector, it is final to send out
It send instruction control high-speed holder motor to be rotated, the central region of high-definition camera is moved to target position.In addition, high definition
Video camera can distance and vector according to mobile target relative to high-definition cameraLength carry out zoom, it is ensured that pass through
High-definition camera function clearly takes the clear image of mobile target.
In addition, the foreign bodies detection module of high-definition tripod video camera realize invasion foreign matter real-time detection method, below it is right
Invasion foreign matter real-time detection method is described in detail:
For arbitrary rail track bad border, possible invasion foreign matter target be mainly the resident of Along Railway, vehicle,
House refuse etc., or pass through or rest on the various animals on rail track, and the branch due to caused by adverse circumstances, fall
Stone, mud-rock flow etc., the more seriously malicious sabotage of unauthorized person such as damage railway equipment, equipment and on railway track
Foreign matter is placed to threaten train operating safety.Which kind of either above-mentioned situation, the generation of foreign matter certainly exist one and dynamically enter
Invade process.Therefore, in the present embodiment invasion foreign matter real-time detection method from analysis foreign matter target dynamic phagocytic process,
Type, size, position, the extent of injury etc. of foreign matter can be analyzed during foreign body intrusion, so as to
Early warning is carried out to it and is removed at the first time, has high timeliness.
Since rail track is relatively limited region, and for foreign matter target, its intrusion direction be should be from railway guardrail
The orbital position in rail track center is gradually advanced to outside.And the detection microwave that microwave radar is sent can cover route guardrails of both sides
Interior all areas form one of height up to 10 meters of even higher radar microwave walls in whole rail track overhead.
Therefore, no matter foreign matter is invaded from which direction, including very unexpectedly, can all be arrived by microwave radar equipment sensing, then be driven immediately
Dynamic high-definition tripod video camera goes to foreign matter target position, shoots the real-time high definition image of foreign matter, finally carries out type, the material of foreign matter
Matter, size analysis etc..6 its specific implementation steps is described in detail with reference to the accompanying drawing:
Step A: for the mobile target detected at any time by radar equipment, such as Y1, Y2, W1 in Fig. 3, if its
The position coordinates detected by radar are W1 (X, Y, Z, V), which is using ground level as X/Y plane, rail track side
To for Y-axis, corresponding vertical direction is X-axis, and the direction perpendicular to ground level is Z axis, the throwing of radar center position on the ground
Shadow position is the coordinate values under the three-dimensional system of coordinate of origin, the movement speed of numerical value V time shift moving-target thus.High-definition tripod is taken the photograph
The central region of video camera is directed at the target, and every by the horizontal and vertical angle position of rotation high-speed holder by camera
An image taking is carried out every time T, picture two is shot every time and opens, every interval time isUntil the target stops movement,
Or having moved out the reconnaissance range of microwave radar equipment, i.e., the target is in time K0It is no longer detected by radar in second.If
Target is moved in the monitoring range of microwave radar always, then is continuously detected K in the target1Start to be walked after second
Rapid B.Wherein time K0,K1It can be set according to field condition, as the present embodiment is set as 5 seconds.And between the time shot every time
It can be calculated according to the movement speed V of target every T, specific calculation formula is as follows:
Wherein, V is the movement velocity of mobile target, and unit is meter per second；T is image capturing time interval, and unit is milli
Second, T is up to 1 second, and minimum 100 milliseconds.
2K high-definition images of step B: analytical procedure A shooting, wherein K is the number of shooting, calculates the true ruler of target
Very little size.
Step B1: carrying out gray processing for two pictures shot every time and do difference operation, due to two pictures shooting when
Between be spaced smaller, therefore its background frame is almost the same when shooting.Therefore, two pictures of shooting are carrying out gray processing and are doing every time
After difference, it can only be had differences in mobile target position.After carrying out morphologic expansion process to the picture after making the difference again, i.e.,
It can obtain mobile specific location and its chamfered shape of the target on picture.
Step B2: according to the zoom multiple of camera lens when the lens parameters and picture shooting of high-definition camera, when obtaining shooting
The focal length F of camera lens.Look-up table can be used in specifically used method, i.e., in advance according to the lens parameters of high-definition camera, by video camera
Zoom multiple and corresponding focal length it is tabulating, when video camera when shooting, first obtain the zoom multiple of video camera at this time, then
The focal length of camera lens at this time can be obtained according to this table.
Step B3: according to video camera Gaussian imaging equation and picture shooting time shift moving-target and video camera distance D, meter
Calculate the size of the image distance L of mobile object at this time.Calculation formula is as follows:
Step B4: the mobile target obtained according to the ratio of object distance D and image distance L and step B1 profile on high-definition image
Size, can calculate this time shooting mobile target true sectional area size.Calculation formula is as follows:
Wherein, W is the true cross-sectional width of mobile target, and H is the true depth of section of mobile target, and w is mobile target
Width on the image, h are the height of mobile target on the image.
Step C: the mobile Target Coordinate Data detected according to K times of the microwave radar equipment of step A, and in step A institute
It states and identifies every time the position coordinates that mobile target is detected in coordinate system, as shown in Figure 5.And for there is multiple targets simultaneously
By microwave radar equipment sensing to the case where, then can be according to the position coordinates and movement speed that the mobile target last time be detected
For degree to judge whether it is the same target, specific judgment formula is as follows:
Wherein, P0It is to be detected position, P the last time for mobile target1It is present position, V for mobile target0For
Movement velocity when the mobile target last time is detected, T1、T0Specific moment when being detected for mobile target.
Step D: the moving object position coordinate data identified according to step C carries out curve fitting, obtains the target
Space motion path equation.
Step E: it according to the space motion path equation for the mobile target that previous step is calculated, and combines in step A
The termination condition that picture is continuously shot, to determine the concrete type of target.
Step E1: if mobile target has moved out the monitoring range of radar equipment, which only passes through iron
Road land, such as aerial flying bird, the animal across rail track will not be to column if not having train to pass through at this time
Vehicle operational safety impacts, then can delete the movement target from system.
Step E2: if mobile target has stopped moving, and its curvilinear equation is parabola, then according to parabolic curve equation
It calculates it and moves the final landing position of target.As shown in Figure 5.If landing position outside the monitoring range of microwave radar equipment,
Then it does not have risk equally, can also exclude the target.If landing position in monitoring range, carries out step F.
Step E3: if mobile target Continuous moves in microwave radar monitoring range, which has risk,
It needs further to analyze its concrete type, i.e. progress step F.
The final position of mobile target: being placed in the central region of high-definition tripod video camera by step F first, and according to
Mobile target size size obtained in step B carries out zoom, in order to which high-definition tripod video camera can just see the whole of target
A profile.The high definition contour images for moving target at this time are shot again.Image is carried out further according to the profile and texture of mobile target
It with analysis, i.e., is matched in the image data base of system, so that the concrete type of the target can be obtained.Wherein image data
Library can configured in advance, can also be according to the mobile target image clapped every time, to update system image data library, image data base
In foreign matter target type data it is more, then the concrete type for the foreign matter target that can more subtly classify.Wherein image data base
The image classification method based on deep learning can be used to carry out in acquisition, training and update.
Deep learning is a kind of based on the method for carrying out representative learning to data in machine learning.This method is relative to tradition
Image classification and recognition methods for, be all greatly improved on accuracy of identification and recognition efficiency.Because this method is not
The global characteristics of image have been concerned only with, very important image local feature when image recognition is even more taken full advantage of.And it is any
Two objects are just finally presented as the otherness of individual due to the difference of the local feature data at its each position.Therefore, will scheme
The precision of identification can be greatly improved as the analysis and matching of local feature have been fused in image recognition algorithm.Based on depth
The image classification of habit and the advantage of recognition methods are exactly to be to hold image overall and its comprehensive of local feature.And this office
Portion's relevance is characterized in what other recognizers can not be extracted.It describes as piece image can be used various features, both may be used
To analyze the variation of its color on the whole, variation tendency as analyzed each pixel color component on image can also be
Image abstraction it is expressed as a series of lines, combination of specific shape etc..And each such character representation is known as the image
A dimension, the dimension used is more, and final accuracy of identification is higher.Its algorithmic procedure and basic step are as follows:
Step F1: the high-definition image of all angles of various possible invasion foreign matters is collected.Foreign matter can such as be divided in advance
For various types such as the mankind, meiofauna, larger animal, falling rocks, branches, the photograph of each type foreign matter all angles is then shot
The method of piece is collected.Real-time update can also be carried out in actual scene use, i.e., when present system finds one kind
When new foreign matter, the foreign matter image update of real-time scene into image data base and is subjected to new type label automatically, with system
Continuous operation, image data base also constantly improve, precisely.
Step F2: pre-processing image data.Preliminary treatment is carried out to all images in the image data base being collected into, such as
Carry out the operation such as gray processing, mean normalization, feature extraction.
Step F3: deep learning model will be added by pretreated picture, and will carry out model training.
Step F4: deep learning training result is extracted.
Step F5: it when increasing new foreign matter image or new foreign matter type, repeats the above steps.
Step G: according to the type of foreign matter target, size, final position etc., to calculate the danger of foreign matter target
Coefficients R.Calculation formula is as follows:
Wherein, R0For danger coefficient corresponding to certain foreign matter type, i.e. the type foreign matter thing in system image data library
The danger coefficient first configured, S are the sectional area size of foreign matter, and D is foreign matter size at a distance from rail track rail center.
Step H: according to the danger coefficient R of foreign matter, foreign matter is divided into general warning, removes, excludes immediately in limited time etc. and is pre-
Alert rank, ultimately generates and issues warning information.
The above is the preferred embodiment of the present invention, it will be understood by a person skilled in the art that the present invention is not by above-mentioned reality
The limitation for applying example, do not depart from basic principle of the invention, primary structure, using territory and using purpose under the premise of,
The present invention also has various similar changes and improvements, these changes and improvements all should belong within protection scope of the present invention.
Claims (10)
1. a kind of rail track foreign matter detecting method based on radar and image analysis, which comprises the steps of:
(1) by being deployed in the microwave radar equipment in rail track roadside, whether there is mobile mesh in its monitoring range of real-time detection
Mark, and send the position coordinates of all mobile targets and velocity information in corresponding high-definition tripod video camera；
(2) high-definition tripod video camera calculates mobile target relative to height according to the relative position of itself and microwave radar equipment
The position of clear video camera, adjusts shooting direction and focal length, is continuously shot mobile target image, up to target stopping movement or
Move out the reconnaissance range of microwave radar equipment；
(3) foreign bodies detection is carried out according to the realtime graphic of high-definition tripod video camera shooting, comprising:
The position moved target on the image and chamfered shape are obtained using the image shot every time；
The full-size(d) of mobile target is calculated according to video camera imaging principle；
It is carried out curve fitting according to the position coordinates of mobile target and obtains the space motion path equation of mobile target；
The type of target is determined in conjunction with the termination condition that the space motion path equation and picture of mobile target are continuously shot, if moving
The monitoring range or target landing position that moving-target has moved out radar equipment are outside the monitoring range of radar equipment, then it is assumed that
Target does not have risk, excludes target；If target Continuous moves in microwave radar monitoring range or landing position is monitoring
In range, then images match analysis is carried out according to the profile of target and texture, obtain foreign matter type and calculate the danger of foreign matter target
Dangerous coefficient；
(4) Warning Service device is sent by foreign matter parameter, danger coefficient and realtime graphic, Warning Service device is according to predefined pre-
Police regulations then, generate and send early warning order to corresponding prior-warning device or the platform of docking.
2. a kind of rail track foreign matter detecting method based on radar and image analysis according to claim 1, feature
It is, in the step (3), by carrying out gray processing operation to two images being continuously shot every time and make difference operation, then into
Capable morphologic expansion process, obtains the position moved target on the image and chamfered shape.
3. a kind of rail track foreign matter detecting method based on radar and image analysis according to claim 1, feature
It is, in the step (3), the full-size(d) calculation formula of mobile target are as follows:
Wherein, W is the true cross-sectional width of mobile target, and H is the true depth of section of mobile target, and w is that mobile target is being schemed
As upper width, h is the height of mobile target on the image, and D is shooting time shift moving-target at a distance from video camera, and L is movement
The image distance of target.
4. a kind of rail track foreign matter detecting method based on radar and image analysis according to claim 1, feature
Be, in the step (3), for have simultaneously multiple targets by microwave radar equipment sensing to the case where, according to mobile target
Position coordinates that last time is detected and movement velocity judge whether it is the same target, specific judgment formula are as follows:
Wherein, P0It is to be detected position, P the last time for mobile target1It is present position, V for mobile target0For movement
Movement velocity when the target last time is detected, T1、T0Specific moment when being detected for mobile target.
5. a kind of rail track foreign matter detecting method based on radar and image analysis according to claim 1, feature
It is, in the step (3), described image the matching analysis uses the image classification method based on deep learning, each by collecting
Deep learning model is added in the image that kind may invade all angles of foreign matter, carries out model training, obtains disaggregated model；Finally
By the contour images input disaggregated model of the mobile target taken to obtain foreign matter type.
6. a kind of rail track foreign matter detecting method based on radar and image analysis according to claim 1, feature
It is, in the step (3), the calculation formula of the danger coefficient R of foreign matter target are as follows:
Wherein, R0Danger coefficient corresponding to foreign matter type for configured in advance, S be foreign matter sectional area size, D be foreign matter with
Rail track rail center apart from size.
7. a kind of rail track foreign matter detection system based on radar and image analysis characterized by comprising be mounted on railway
The microwave radar equipment in route roadside is mounted on the high-definition tripod camera shooting corresponding with microwave radar equipment of rail track roadside
Machine and Warning Service device；
The microwave radar equipment, for detecting in its reconnaissance range with the presence or absence of mobile target, and mobile target speed,
The three-dimensional coordinate information of position, and will send information to corresponding high-definition tripod video camera；
The high-definition tripod video camera, for receive microwave radar equipment transmission moving target position after, according to itself
With the relative position of microwave radar equipment, calculate position of the mobile target relative to high-definition camera, adjustment shooting direction and
Focal length is continuously shot mobile target image, until target stops reconnaissance range that is mobile or having moved out microwave radar equipment；
And foreign bodies detection is carried out by realtime graphic of the built-in foreign bodies detection module to shooting, pass through network after detecting foreign matter
Foreign matter warning information is sent to Warning Service device, wherein foreign matter detecting method is included: and obtained to move using the image shot every time
Moving-target position on the image and chamfered shape；The full-size(d) of mobile target is calculated according to video camera imaging principle；According to
The position coordinates of mobile target, which carry out curve fitting, obtains the space motion path equation of mobile target；In conjunction with the sky of mobile target
Between the termination condition that is continuously shot of Movement Locus Equation and picture determine the type of target, if mobile target has moved out radar
The monitoring range or target landing position of equipment are outside the monitoring range of radar equipment, then it is assumed that target does not have risk, row
Except target；If target Continuous moves in microwave radar monitoring range or landing position is in monitoring range, according to target
Profile and texture carry out images match analysis, obtain foreign matter type and calculate the danger coefficient of foreign matter target；
The Warning Service device generates early warning order and is sent to installation for receiving foreign matter warning information, and according to early warning rule
In the prior-warning device in rail track roadside or the platform of docking.
8. the rail track foreign matter detection system according to claim 7 based on radar and image analysis, which is characterized in that
The high-definition tripod video camera, microwave radar equipment and prior-warning device are fixed in the upright bar installed by rail track, described vertical
It is also equipped on bar and provides the outer of power supply and network communication for the high-definition tripod video camera, microwave radar equipment and prior-warning device
Peripheral equipment.
9. the rail track foreign matter detection system according to claim 7 based on radar and image analysis, which is characterized in that
It include high-definition tripod video camera, microwave radar equipment and early warning dress by being installed a set of by rail track every set distance L
The equipment set completely monitors a rail track, wherein distance L monitors length and high-definition camera according to microwave radar maximum
Maximum sighting distance determines.
10. the rail track foreign matter detection system according to claim 7 based on radar and image analysis, feature exist
In the reconnaissance range length of the microwave radar equipment is configured to the distance of microwave radar equipment adjacent thereto, and width configuration is
The distance between rail track guardrails of both sides.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810978155.0A CN109164443A (en) | 2018-08-27 | 2018-08-27 | Rail track foreign matter detecting method and system based on radar and image analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810978155.0A CN109164443A (en) | 2018-08-27 | 2018-08-27 | Rail track foreign matter detecting method and system based on radar and image analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109164443A true CN109164443A (en) | 2019-01-08 |
Family
ID=64896908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810978155.0A Pending CN109164443A (en) | 2018-08-27 | 2018-08-27 | Rail track foreign matter detecting method and system based on radar and image analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109164443A (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109752731A (en) * | 2019-01-10 | 2019-05-14 | 北京佳讯飞鸿电气股份有限公司 | A kind of one-way revolving door foreign matter detecting method |
| CN109816702A (en) * | 2019-01-18 | 2019-05-28 | 苏州矽典微智能科技有限公司 | A kind of multiple target tracking device and method |
| CN109885560A (en) * | 2019-02-18 | 2019-06-14 | 中国石油大学(北京) | The cleaning method and device of shale gas pressure break equipment infrared imaging inspection data |
| CN110058239A (en) * | 2019-04-29 | 2019-07-26 | 上海保隆汽车科技股份有限公司 | A kind of vehicle-mounted millimeter wave radar installations and object detection method |
| CN110146899A (en) * | 2019-05-28 | 2019-08-20 | 许灵男 | A kind of terminal radar barrier sensory perceptual system |
| CN110853307A (en) * | 2019-10-18 | 2020-02-28 | 深圳供电局有限公司 | Laser monitoring method and monitoring device for power transmission channel external broken environment |
| CN110989026A (en) * | 2019-11-16 | 2020-04-10 | 南京派光智慧感知信息技术有限公司 | Safety monitoring method for cross-railway bridge |
| CN111526352A (en) * | 2020-07-02 | 2020-08-11 | 北京大成国测科技有限公司 | Railway foreign matter anti-invasion three-dimensional intelligent recognition robot equipment |
| CN111798634A (en) * | 2020-06-29 | 2020-10-20 | 杭州海康威视数字技术股份有限公司 | Perimeter detection method and device |
| CN111899447A (en) * | 2019-05-06 | 2020-11-06 | 杭州海康威视数字技术股份有限公司 | Monitoring system and method |
| CN112009525A (en) * | 2020-08-21 | 2020-12-01 | 中科蓝卓(北京)信息科技有限公司 | Track foreign matter intrusion detection system based on remote sensing technology |
| CN112101233A (en) * | 2020-09-16 | 2020-12-18 | 中南大学 | Method and system for monitoring foreign matter invasion of rail transit line and computer storage medium |
| CN112346077A (en) * | 2019-08-08 | 2021-02-09 | 杭州海康威视系统技术有限公司 | Ship superelevation detection method and device, electronic equipment and storage medium |
| CN112365524A (en) * | 2020-11-10 | 2021-02-12 | 宁波博登智能科技有限责任公司 | High-altitude parabolic real-time alarm system based on time sequence images |
| CN112637557A (en) * | 2020-12-09 | 2021-04-09 | 光谷技术有限公司 | Ecological monitoring and early warning method and system |
| CN112702571A (en) * | 2020-12-18 | 2021-04-23 | 福建汇川物联网技术科技股份有限公司 | Monitoring method and device |
| CN112767647A (en) * | 2020-12-29 | 2021-05-07 | 深圳力维智联技术有限公司 | Danger early warning method, device, equipment and computer readable storage medium |
| CN113240887A (en) * | 2021-05-14 | 2021-08-10 | 深圳市安泰数据监测科技有限公司 | Slope rockfall collapse monitoring method, device and equipment |
| CN113313182A (en) * | 2021-06-07 | 2021-08-27 | 北博（厦门）智能科技有限公司 | Target identification method and terminal based on radar and video fusion |
| CN113589305A (en) * | 2021-07-23 | 2021-11-02 | 广东智通睿新智能科技有限公司 | Railway platform clearance foreign matter detecting system |
| CN113947731A (en) * | 2021-12-21 | 2022-01-18 | 成都中轨轨道设备有限公司 | Foreign matter identification method and system based on contact net safety inspection |
| CN114882661A (en) * | 2022-04-29 | 2022-08-09 | 广州小鹏汽车科技有限公司 | Outdoor early warning method, device, system and computer readable storage medium |
| CN114973664A (en) * | 2022-05-17 | 2022-08-30 | 保银信科信息技术(湖北)有限公司 | Machine vision-based urban traffic online intelligent monitoring and early warning management system |
| WO2023279786A1 (en) * | 2021-07-08 | 2023-01-12 | 华为技术有限公司 | Perimeter detection method and apparatus, and device and system |
| CN115892131A (en) * | 2023-02-15 | 2023-04-04 | 深圳大学 | Intelligent monitoring method and system for subway tunnel |
| CN116055338A (en) * | 2023-03-28 | 2023-05-02 | 杭州觅睿科技股份有限公司 | False alarm eliminating method, device, equipment and medium |
| TWI833130B (en) * | 2021-11-12 | 2024-02-21 | 為升電裝工業股份有限公司 | Rail intrusion detection system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102253381A (en) * | 2011-04-20 | 2011-11-23 | 上海交通大学 | System and method for automatically detecting foreign object debris (FOD) on airfield runways |
| CN105303757A (en) * | 2015-09-25 | 2016-02-03 | 凌云光技术集团有限责任公司 | Power grid intelligent monitoring method for preventing artificial external force damage and system thereof |
| CN105959655A (en) * | 2016-07-18 | 2016-09-21 | 四川君逸数码科技股份有限公司 | Alarm method and device for identifying region invasion by intelligent eye |
| CN105989682A (en) * | 2015-02-05 | 2016-10-05 | 国家电网公司 | Safety early warning and monitoring system and monitoring method for construction machinery under power transmission line |
| CN107697102A (en) * | 2017-09-01 | 2018-02-16 | 南京理工大学 | Information gathering, identification and the caution system of high-speed railway foreign body intrusion |
| CN107728136A (en) * | 2017-11-29 | 2018-02-23 | 航科院（北京）科技发展有限公司 | A kind of airfield runway monitoring foreign bodies and removing guiding system and method |
| CN107911659A (en) * | 2017-11-20 | 2018-04-13 | 重庆交通职业学院 | Railroad track section inspection warning system on duty |
| KR20180061929A (en) * | 2016-11-30 | 2018-06-08 | 주식회사 모디엠 | MOBILE 3D MAPPING SYSTEM OF RAILWAY FACILITIES EQUIPPED WITH DUAL LIDAR and 3D MAPPING METHOD USING THE SAME |
-
2018
- 2018-08-27 CN CN201810978155.0A patent/CN109164443A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102253381A (en) * | 2011-04-20 | 2011-11-23 | 上海交通大学 | System and method for automatically detecting foreign object debris (FOD) on airfield runways |
| CN105989682A (en) * | 2015-02-05 | 2016-10-05 | 国家电网公司 | Safety early warning and monitoring system and monitoring method for construction machinery under power transmission line |
| CN105303757A (en) * | 2015-09-25 | 2016-02-03 | 凌云光技术集团有限责任公司 | Power grid intelligent monitoring method for preventing artificial external force damage and system thereof |
| CN105959655A (en) * | 2016-07-18 | 2016-09-21 | 四川君逸数码科技股份有限公司 | Alarm method and device for identifying region invasion by intelligent eye |
| KR20180061929A (en) * | 2016-11-30 | 2018-06-08 | 주식회사 모디엠 | MOBILE 3D MAPPING SYSTEM OF RAILWAY FACILITIES EQUIPPED WITH DUAL LIDAR and 3D MAPPING METHOD USING THE SAME |
| CN107697102A (en) * | 2017-09-01 | 2018-02-16 | 南京理工大学 | Information gathering, identification and the caution system of high-speed railway foreign body intrusion |
| CN107911659A (en) * | 2017-11-20 | 2018-04-13 | 重庆交通职业学院 | Railroad track section inspection warning system on duty |
| CN107728136A (en) * | 2017-11-29 | 2018-02-23 | 航科院（北京）科技发展有限公司 | A kind of airfield runway monitoring foreign bodies and removing guiding system and method |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109752731A (en) * | 2019-01-10 | 2019-05-14 | 北京佳讯飞鸿电气股份有限公司 | A kind of one-way revolving door foreign matter detecting method |
| CN109816702A (en) * | 2019-01-18 | 2019-05-28 | 苏州矽典微智能科技有限公司 | A kind of multiple target tracking device and method |
| CN109885560A (en) * | 2019-02-18 | 2019-06-14 | 中国石油大学(北京) | The cleaning method and device of shale gas pressure break equipment infrared imaging inspection data |
| CN110058239A (en) * | 2019-04-29 | 2019-07-26 | 上海保隆汽车科技股份有限公司 | A kind of vehicle-mounted millimeter wave radar installations and object detection method |
| CN110058239B (en) * | 2019-04-29 | 2023-08-01 | 上海保隆汽车科技股份有限公司 | Vehicle-mounted millimeter wave radar device and target detection method |
| CN111899447A (en) * | 2019-05-06 | 2020-11-06 | 杭州海康威视数字技术股份有限公司 | Monitoring system and method |
| CN110146899A (en) * | 2019-05-28 | 2019-08-20 | 许灵男 | A kind of terminal radar barrier sensory perceptual system |
| CN112346077A (en) * | 2019-08-08 | 2021-02-09 | 杭州海康威视系统技术有限公司 | Ship superelevation detection method and device, electronic equipment and storage medium |
| CN110853307A (en) * | 2019-10-18 | 2020-02-28 | 深圳供电局有限公司 | Laser monitoring method and monitoring device for power transmission channel external broken environment |
| CN110853307B (en) * | 2019-10-18 | 2022-01-04 | 深圳供电局有限公司 | Laser monitoring method and monitoring device for power transmission channel external broken environment |
| CN110989026A (en) * | 2019-11-16 | 2020-04-10 | 南京派光智慧感知信息技术有限公司 | Safety monitoring method for cross-railway bridge |
| CN111798634A (en) * | 2020-06-29 | 2020-10-20 | 杭州海康威视数字技术股份有限公司 | Perimeter detection method and device |
| CN111798634B (en) * | 2020-06-29 | 2022-02-01 | 杭州海康威视数字技术股份有限公司 | Perimeter detection method and device |
| CN111526352B (en) * | 2020-07-02 | 2020-10-23 | 北京大成国测科技有限公司 | Railway foreign matter anti-invasion three-dimensional intelligent recognition robot equipment |
| CN111526352A (en) * | 2020-07-02 | 2020-08-11 | 北京大成国测科技有限公司 | Railway foreign matter anti-invasion three-dimensional intelligent recognition robot equipment |
| CN112009525A (en) * | 2020-08-21 | 2020-12-01 | 中科蓝卓(北京)信息科技有限公司 | Track foreign matter intrusion detection system based on remote sensing technology |
| CN112101233A (en) * | 2020-09-16 | 2020-12-18 | 中南大学 | Method and system for monitoring foreign matter invasion of rail transit line and computer storage medium |
| CN112101233B (en) * | 2020-09-16 | 2024-04-09 | 中南大学 | Rail transit line foreign matter intrusion monitoring method, system and computer storage medium |
| CN112365524B (en) * | 2020-11-10 | 2024-03-29 | 宁波博登智能科技有限公司 | High-altitude parabolic real-time alarm system based on time sequence image |
| CN112365524A (en) * | 2020-11-10 | 2021-02-12 | 宁波博登智能科技有限责任公司 | High-altitude parabolic real-time alarm system based on time sequence images |
| CN112637557A (en) * | 2020-12-09 | 2021-04-09 | 光谷技术有限公司 | Ecological monitoring and early warning method and system |
| CN112702571A (en) * | 2020-12-18 | 2021-04-23 | 福建汇川物联网技术科技股份有限公司 | Monitoring method and device |
| CN112767647A (en) * | 2020-12-29 | 2021-05-07 | 深圳力维智联技术有限公司 | Danger early warning method, device, equipment and computer readable storage medium |
| CN113240887A (en) * | 2021-05-14 | 2021-08-10 | 深圳市安泰数据监测科技有限公司 | Slope rockfall collapse monitoring method, device and equipment |
| CN113313182B (en) * | 2021-06-07 | 2023-10-17 | 北博（厦门）智能科技有限公司 | Target identification method and terminal based on radar and video fusion |
| CN113313182A (en) * | 2021-06-07 | 2021-08-27 | 北博（厦门）智能科技有限公司 | Target identification method and terminal based on radar and video fusion |
| WO2023279786A1 (en) * | 2021-07-08 | 2023-01-12 | 华为技术有限公司 | Perimeter detection method and apparatus, and device and system |
| CN113589305A (en) * | 2021-07-23 | 2021-11-02 | 广东智通睿新智能科技有限公司 | Railway platform clearance foreign matter detecting system |
| TWI833130B (en) * | 2021-11-12 | 2024-02-21 | 為升電裝工業股份有限公司 | Rail intrusion detection system |
| CN113947731B (en) * | 2021-12-21 | 2022-07-22 | 成都中轨轨道设备有限公司 | Foreign matter identification method and system based on contact net safety inspection |
| CN113947731A (en) * | 2021-12-21 | 2022-01-18 | 成都中轨轨道设备有限公司 | Foreign matter identification method and system based on contact net safety inspection |
| CN114882661A (en) * | 2022-04-29 | 2022-08-09 | 广州小鹏汽车科技有限公司 | Outdoor early warning method, device, system and computer readable storage medium |
| CN114973664A (en) * | 2022-05-17 | 2022-08-30 | 保银信科信息技术(湖北)有限公司 | Machine vision-based urban traffic online intelligent monitoring and early warning management system |
| CN115892131A (en) * | 2023-02-15 | 2023-04-04 | 深圳大学 | Intelligent monitoring method and system for subway tunnel |
| CN116055338A (en) * | 2023-03-28 | 2023-05-02 | 杭州觅睿科技股份有限公司 | False alarm eliminating method, device, equipment and medium |
| CN116055338B (en) * | 2023-03-28 | 2023-08-11 | 杭州觅睿科技股份有限公司 | False alarm eliminating method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109164443A (en) | Rail track foreign matter detecting method and system based on radar and image analysis | |
| KR101533905B1 (en) | A surveillance system and a method for detecting a foreign object, debris, or damage in an airfield | |
| CN110889350B (en) | Line obstacle monitoring and alarming system and method based on three-dimensional imaging | |
| CN104971458B (en) | Based on the more fire source recognition methods for automatically tracking positioning jet stream extinguishing device | |
| Semertzidis et al. | Video sensor network for real-time traffic monitoring and surveillance | |
| CN101465033B (en) | Automatic tracking recognition system and method | |
| CN107728136A (en) | A kind of airfield runway monitoring foreign bodies and removing guiding system and method | |
| CN201278180Y (en) | Automatic tracking recognition system | |
| CN108648456A (en) | A kind of traffic events monitoring device and method based on video and radar | |
| CN104613892A (en) | Video detection technology and laser ranging technology integrated compound snow depth monitoring system | |
| CN105389921A (en) | Airport runway foreign body monitoring system and method | |
| CN107360394A (en) | More preset point dynamic and intelligent monitoring methods applied to frontier defense video monitoring system | |
| WO2023221425A1 (en) | Ship lockage safety detection method | |
| CN105222752B (en) | A kind of portable road face detection means and method using structure light | |
| CN105303844A (en) | Night highway agglomerate fog automatic detection device on the basis of laser and detection method thereof | |
| CN112349055A (en) | Target monitoring device and method based on radar and video linkage | |
| CN116453278A (en) | Intrusion monitoring method combining deep learning intelligent detection and optical fiber vibration sensing | |
| CN112257683A (en) | Cross-mirror tracking method for vehicle running track monitoring | |
| CN114283544B (en) | Railway platform intrusion monitoring system and method based on artificial intelligence | |
| CN207502720U (en) | A kind of airfield runway monitoring foreign bodies and removing guiding system | |
| KR101542134B1 (en) | The apparatus and method of surveillance a rock fall based on smart video analytic | |
| CN115755069A (en) | Track foreign matter detection device, detection system and detection method | |
| CN112537302B (en) | Driverless traffic vehicle lane keeping method and device and traffic vehicle | |
| KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN113740847A (en) | Multi-radar cooperative detection alarm system based on humanoid target recognition |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190108 |
|
| RJ01 | Rejection of invention patent application after publication |