CN101626493A - Method for judging forward motion direction of pedestrian by combining laser scanning and videos - Google Patents
Method for judging forward motion direction of pedestrian by combining laser scanning and videos Download PDFInfo
- Publication number
- CN101626493A CN101626493A CN200910090703A CN200910090703A CN101626493A CN 101626493 A CN101626493 A CN 101626493A CN 200910090703 A CN200910090703 A CN 200910090703A CN 200910090703 A CN200910090703 A CN 200910090703A CN 101626493 A CN101626493 A CN 101626493A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- laser scanning
- data
- video image
- video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 8
- 230000036541 health Effects 0.000 claims description 3
- 241001269238 Data Species 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention provides a method for judging a forward motion direction of a pedestrian by combining laser scanning and videos, which comprises the following steps: (a) carrying out vertical laser scanning and video shooting on pedestrians overhead to acquire laser scanning data and video images; (b) analyzing the laser scanning data to know that a pedestrian is scanned by laser in a frame of the laser scanning data and obtain the analytical data of the frame of the laser scanning data; (c) detecting the video images according to the analytical data so as to find out the body characteristics of the pedestrian; and (d) tracking the pedestrian in the video images according to the body characteristics so as to judge the forward motion direction of the pedestrian. The method can accurately and effectively detect a human body target in the videos by combining the laser scanning and the videos so as to carry out tracking processing and has small influence of external environment.
Description
Technical field
The present invention relates to the pedestrian advancing direction judging method of a kind of combining laser scanning and video.
Background technology
At present, along with the continuous construction of rapid economy development, infrastructure, extensive frequent day by day the holding of organization activity, mobile crowded place, increasing such as the flow of personnel amount that public transport, subway, airport, railway station etc. are local.Meanwhile, detect the passenger flow information that is obtained by passenger flow and also become increasingly complex, amount of information is also huge day by day.Passenger flow information is a significant data of weighing operation situation, by the statistical analysis to passenger flow information, not only can obtain the passenger flow operation conditions of this traffic zone, and can utilize the detection data effectively to organize operation, scheduling to arrange work.Simultaneously,, also good forewarning function can be played, in case of emergency emergency measures can also be optimized for the bigger zone of people's current density.
In passenger flow statistics, often be unable to do without statistics, and the passenger flow direction is presented as passenger/pedestrian's direction of advance on individuality to the passenger flow direction.Therefore, be a very important job during passenger flow statistics is analyzed to the judgement of pedestrian advancing direction.Prior art generally is to adopt the method for video tracking to judge pedestrian's direction of advance to the judgement of pedestrian advancing direction, this class technology often exists the influence of situations such as being subject to background, weather, shelter, shade, and video tracking often needs at first to determine the initial position of human body, this need carry out thorough search to the video image of certain frame, and more complicated and efficient are not high.
Therefore, need a kind of pedestrian advancing direction judging method that can overcome above defective.
Summary of the invention
The invention provides the determination methods of the pedestrian advancing direction of a kind of combining laser scanning that can overcome the above problems and video.
In first aspect, the invention provides the pedestrian advancing direction judging method of a kind of combining laser scanning and video, comprising: step a, carry out vertical laser scanning and video capture from the top to the pedestrian, obtain laser scanning data and video image; Step b analyzes described laser scanning data, and learning in a certain frame laser scanning data has the pedestrian to be arrived by laser scanning, and draws the analysis data of this frame laser scanning data; Step c detects in video image according to described analysis data, finds described pedestrian's characteristics of human body; Steps d is followed the tracks of it in video image according to described characteristics of human body, judges described pedestrian's direction of advance.
In second aspect, the invention provides the pedestrian advancing direction judgment means of a kind of combining laser scanning and video, comprising: module, be used for carrying out vertical laser scanning and video capture from the top to the pedestrian, obtain laser scanning data and video image; Module is used for described laser scanning data is analyzed, and learning in a certain frame laser scanning data has the pedestrian to be arrived by laser scanning, and draws the analysis data of this frame laser scanning data; Module is used for detecting at video image according to described analysis data, finds described pedestrian's characteristics of human body; Module is used for it being followed the tracks of at video image according to described characteristics of human body, judges described pedestrian's direction of advance.
The present invention can detect human body target accurately and effectively, thereby follow the tracks of processing by combining laser scanning and video in video, be subjected to the influence of external environment little.
Description of drawings
Below with reference to accompanying drawings specific embodiments of the present invention is described in detail, in the accompanying drawings:
Fig. 1 is the schematic diagram of judgement pedestrian advancing direction according to an embodiment of the invention.
Embodiment
Method of the present invention is, from passenger flow passage top it is carried out vertical laser scanning and video capture, time and position according to the pedestrian during by laser scanning face, really allocation carries out characteristics of human body's detection in a frame video image, if arrive the characteristics of human body in this position probing, then in continuous video frames, follow the tracks of this characteristics of human body, judge its direction of advance.
Fig. 1 is the schematic diagram of judgement pedestrian advancing direction according to an embodiment of the invention.
As shown in the figure, carry out vertical laser scanning and video capture, obtain laser scanning data and video stream data respectively from passenger flow passage top.The laser scanning data of each frame and video stream data all are stored for carrying out data analysis.
The laser scanning initial point apart from the vertical height on ground generally more than 2 meters.In general, laser scanning module is launched by the point by point scanning mode and is received laser beam, thereby obtains the laser scanning data of each frame.Laser beam sweep limits on the ground is starting point and the terminal point of this section in the ground projection, and described starting point and terminal point can freely be set, and can set a plurality of.Specifically, laser scanning data comprises distance, sweep time and the scanning frequency of the scan origin of scanned object range sweep module.Laser scanning data and manikin storehouse are compared, can draw the profile information of human body, comprise height, width of human body or the like.
Video stream data is that current scene is taken, and obtains the video image of continuous multiple frames.
Should be pointed out that the video capture zone should comprise the laser scanning scope interior, promptly the pedestrian by being arrived by video capture in the laser scanning face.
For the ease of association, need to set up two coordinate systems with regard to laser scanning data and video image respectively in advance.On the one hand, be transverse axis with the intersecting lens on laser scanning face and ground, its left end point is that initial point is set up first coordinate system.Hence one can see that, and as pedestrian during by laser scanning face, its coordinate in first coordinate system for example is (x, 0).On the other hand, position by video capture and laser scanning concerns as can be known, laser scanning face is rendered as a horizontal line section with the intersecting lens on ground in video image, with it is transverse axis, its left end point is that initial point is set up second coordinate system, and hence one can see that, as pedestrian by laser scanning face the time, its coordinate in second coordinate system for example is (x ', 0).
Should be appreciated that the position relation that depends on laser scanning and video camera, the foundation of first and second coordinate systems can have various forms, and should not be confined to above-mentioned example.
Specifically describe below and how in real process, pedestrian advancing direction to be judged.
When finding the pedestrian is arranged by laser scanning face, analyze by laser scanning data present frame, can learn the time t of present frame
1, this pedestrian is at the abscissa x of described first coordinate system
1, the height h
1, and even pedestrian head width w
hWith health width w
bOr the like.
Close on t in the time
1A frame video image in carry out the characteristics of human body and search, preferably, be the time near t
1A frame video image in search.For example, the time of continuous 3 frame laser scanning datas is respectively 8:13:580,8:13:620,8:13:660, and the time of continuous 3 frame video datas is respectively 8:13:590,8:13:630,8:13:670, and wherein a second number is a unit with millisecond ms.If at t
1' constantly, promptly detect the pedestrian in the frame laser scanning data of 8:13:620, then should be at for example t
2' constantly, promptly search in the frame video image of 8:13:630.
At first, in video frame image, determine one and search the position.At this moment, need be about to x with the pedestrian's Coordinate Conversion in first coordinate system to second coordinate system
1Convert x to
1', x
1'=f (x
1, h
1), wherein, x
1' be the abscissa of pedestrian in second coordinate system, x
1Be the abscissa of pedestrian in first coordinate system, h
1Be pedestrian's height, function f () can preestablish according to the position relation of laser scanning and video capture.Then, search in video frame image according to the coordinate after the conversion.The characteristics of human body searches and can adopt various existing Human Detection.
If detect characteristics of human body's (generally being head feature), can determine that then this feature belongs to the aforementioned pedestrian who is scanned in this coordinate points.In another example, can also be according to the pedestrian head width w that analysis draws to laser scanning data
hWith health width w
b, compare with the characteristics of human body who finds, verify whether this characteristics of image belongs to the pedestrian who is arrived by laser scanning.If the comparing result of the two is in certain error permissible range, can judge that then this characteristics of image belongs to the aforementioned pedestrian who is arrived by laser scanning.
Next, in video streaming image, detected characteristics of human body is followed the tracks of.Can perhaps in the multiframe of front and back this characteristics of human body be followed the tracks of at adjacent 2 frames of this video frame image, determine this pedestrian's direction of advance.For example, at t
1' constantly a back frame video image, i.e. t
2' in constantly the frame video image, with (x
1', y
1') coordinate is central point, and a very little region of search is set, search for therein.If at (x
2', y
2') the coordinate place detects identical or approximate characteristics of human body, then relatively before and after characteristics of human body's coordinate points ordinate y of two frames
1' and y
2', promptly know this pedestrian's direction of advance.
Obviously, under the prerequisite that does not depart from true spirit of the present invention and scope, the present invention described here can have many variations.Therefore, the change that all it will be apparent to those skilled in the art that all should be included within the scope that these claims contain.The present invention's scope required for protection is only limited by described claims.
Claims (6)
1. the pedestrian advancing direction judging method of combining laser scanning and video comprises:
Step a carries out vertical laser scanning and video capture from the top to the pedestrian, obtains laser scanning data and video image;
Step b analyzes described laser scanning data, and learning in a certain frame laser scanning data has the pedestrian to be arrived by laser scanning, and draws the analysis data of this frame laser scanning data;
Step c detects in video image according to described analysis data, finds described pedestrian's characteristics of human body;
Steps d is followed the tracks of it in video image according to described characteristics of human body, judges described pedestrian's direction of advance.
2. method according to claim 1, wherein, step c comprises:
According to the pedestrian position in the described analysis data and pedestrian's height, determine the detection position in the video image, carry out characteristics of human body's detection in described detection position.
3. method according to claim 1 also comprises:
According to the pedestrian position in the described analysis data and pedestrian's height, pedestrian's coordinate is transformed into second coordinate system from first coordinate system, and with the coordinate of pedestrian in second coordinate system as the detection position, wherein, described first coordinate system and second coordinate system are to set up according to laser scanning data and video image respectively.
4. method according to claim 1, wherein, step c comprises:
According to the present frame time in the described analysis data, the approaching with it frame video image of select time carries out human detection.
5. method according to claim 1, wherein, step c also comprises:
Verify according to the pedestrian head width in the described analysis data and pedestrian's health width, determine that described characteristics of human body belongs to described pedestrian.
6. the pedestrian advancing direction judgment means of combining laser scanning and video comprises:
Module is used for carrying out vertical laser scanning and video capture from the top to the pedestrian, obtains laser scanning data and video image;
Module is used for described laser scanning data is analyzed, and learning in a certain frame laser scanning data has the pedestrian to be arrived by laser scanning, and draws the analysis data of this frame laser scanning data;
Module is used for detecting at video image according to described analysis data, finds described pedestrian's characteristics of human body;
Module is used for it being followed the tracks of at video image according to described characteristics of human body, judges described pedestrian's direction of advance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100907037A CN101626493B (en) | 2009-08-06 | 2009-08-06 | Method for judging forward motion direction of pedestrian by combining laser scanning and videos |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100907037A CN101626493B (en) | 2009-08-06 | 2009-08-06 | Method for judging forward motion direction of pedestrian by combining laser scanning and videos |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101626493A true CN101626493A (en) | 2010-01-13 |
| CN101626493B CN101626493B (en) | 2012-12-19 |
Family
ID=41522136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100907037A Expired - Fee Related CN101626493B (en) | 2009-08-06 | 2009-08-06 | Method for judging forward motion direction of pedestrian by combining laser scanning and videos |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101626493B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101789077A (en) * | 2010-02-10 | 2010-07-28 | 北京理工大学 | Laser guiding video passenger flow detection method and device |
| CN102306330A (en) * | 2011-08-02 | 2012-01-04 | 北京北大千方科技有限公司 | Laser passenger flow detection method and device |
| CN102622578A (en) * | 2012-02-06 | 2012-08-01 | 中山大学 | Passenger counting system and passenger counting method |
| CN104112305A (en) * | 2013-10-17 | 2014-10-22 | 北京竞业达数码科技有限公司 | Passenger flow monitoring method and apparatus based on dual laser ranging |
| CN104601966A (en) * | 2015-02-09 | 2015-05-06 | 桂林电子科技大学 | Visual analysis visitors flow rate statistical method and visual analysis visitors flow rate statistical equipment based on laser detection lines |
| CN104392522B (en) * | 2014-11-21 | 2016-08-24 | 安徽清新互联信息科技有限公司 | A kind of passenger flow statistical system based on pseudo-depth information |
| WO2018035711A1 (en) * | 2016-08-23 | 2018-03-01 | 深圳市速腾聚创科技有限公司 | Target detection method and system |
| CN110929636A (en) * | 2019-11-20 | 2020-03-27 | 上海融军实业有限公司 | Passenger flow size and direction detection method and system |
| CN112446922A (en) * | 2020-11-24 | 2021-03-05 | 厦门熵基科技有限公司 | Pedestrian reverse judgment method and device for channel gate |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017080A (en) * | 2002-09-30 | 2007-08-15 | 石川岛播磨重工业株式会社 | Device of measuring object |
| CN101216885A (en) * | 2008-01-04 | 2008-07-09 | 中山大学 | Passerby face detection and tracing algorithm based on video |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100537402C (en) * | 2006-06-28 | 2009-09-09 | 上海振华港口机械（集团）股份有限公司 | Container-truck positioning system and method for container crane |
| CN101232571B (en) * | 2008-01-25 | 2010-06-09 | 北京中星微电子有限公司 | Human body image matching method and video analyzing search system |
-
2009
- 2009-08-06 CN CN2009100907037A patent/CN101626493B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017080A (en) * | 2002-09-30 | 2007-08-15 | 石川岛播磨重工业株式会社 | Device of measuring object |
| CN101216885A (en) * | 2008-01-04 | 2008-07-09 | 中山大学 | Passerby face detection and tracing algorithm based on video |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101789077B (en) * | 2010-02-10 | 2012-07-18 | 北京理工大学 | Laser guiding video passenger flow detection method and device |
| CN101789077A (en) * | 2010-02-10 | 2010-07-28 | 北京理工大学 | Laser guiding video passenger flow detection method and device |
| CN102306330A (en) * | 2011-08-02 | 2012-01-04 | 北京北大千方科技有限公司 | Laser passenger flow detection method and device |
| CN102306330B (en) * | 2011-08-02 | 2014-10-29 | 北京北大千方科技有限公司 | Laser passenger flow detection method and device |
| CN102622578A (en) * | 2012-02-06 | 2012-08-01 | 中山大学 | Passenger counting system and passenger counting method |
| CN104112305A (en) * | 2013-10-17 | 2014-10-22 | 北京竞业达数码科技有限公司 | Passenger flow monitoring method and apparatus based on dual laser ranging |
| CN104112305B (en) * | 2013-10-17 | 2016-04-13 | 北京竞业达数码科技有限公司 | Based on passenger flow monitoring method and the device of double excitation range finding |
| CN104392522B (en) * | 2014-11-21 | 2016-08-24 | 安徽清新互联信息科技有限公司 | A kind of passenger flow statistical system based on pseudo-depth information |
| CN104601966A (en) * | 2015-02-09 | 2015-05-06 | 桂林电子科技大学 | Visual analysis visitors flow rate statistical method and visual analysis visitors flow rate statistical equipment based on laser detection lines |
| CN104601966B (en) * | 2015-02-09 | 2017-10-31 | 桂林电子科技大学 | A kind of visual analysis people flow rate statistical method and apparatus based on laser detection line |
| WO2018035711A1 (en) * | 2016-08-23 | 2018-03-01 | 深圳市速腾聚创科技有限公司 | Target detection method and system |
| US10891499B2 (en) | 2016-08-23 | 2021-01-12 | Suteng Innovation Technology Co., Ltd. | Target detection method and system |
| CN110929636A (en) * | 2019-11-20 | 2020-03-27 | 上海融军实业有限公司 | Passenger flow size and direction detection method and system |
| CN112446922A (en) * | 2020-11-24 | 2021-03-05 | 厦门熵基科技有限公司 | Pedestrian reverse judgment method and device for channel gate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101626493B (en) | 2012-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101626493B (en) | Method for judging forward motion direction of pedestrian by combining laser scanning and videos | |
| CN108491758B (en) | Track detection method and robot | |
| US10607090B2 (en) | Train security system | |
| CN103646250B (en) | Pedestrian monitoring method and device based on distance image head and shoulder features | |
| CN102799935B (en) | Human flow counting method based on video analysis technology | |
| Zhao et al. | Detection and tracking of moving objects at intersections using a network of laser scanners | |
| CN102759347B (en) | Online in-process quality control device and method for high-speed rail contact networks and composed high-speed rail contact network detection system thereof | |
| CN101510356B (en) | Video detection system and data processing device thereof, video detection method | |
| CN108229407A (en) | A kind of behavioral value method and system in video analysis | |
| CN101644717B (en) | Pedestrian advancing direction judging method adopting laser scanning | |
| CN101847206A (en) | Pedestrian traffic statistical method and system based on traffic monitoring facilities | |
| CN103714698A (en) | Bus passenger flow statistics system based on distance images | |
| CN105513342A (en) | Video-tracking-based vehicle queuing length calculating method | |
| CN104123776B (en) | A kind of object statistical method based on image and system | |
| CN112113978A (en) | Vehicle-mounted tunnel defect online detection system and method based on deep learning | |
| CN105872970A (en) | Pseudo base station detection system based on street lamp Internet of Things | |
| CN109242035B (en) | Vehicle bottom fault detection device and method | |
| CN105830426A (en) | Video generating method and device of video generating system | |
| CN103646253A (en) | Bus passenger flow statistics method based on multi-motion passenger behavior analysis | |
| Rafique et al. | Optimized real-time parking management framework using deep learning | |
| CN102496275A (en) | Method for detecting overload of coach or not | |
| US11557133B1 (en) | Automatic license plate recognition | |
| CN114627526A (en) | Fusion duplicate removal method and device based on multi-camera snapshot image and readable medium | |
| CN109085177A (en) | A kind of leakage cable appearance detecting method and device based on RFID positioning | |
| US20230365174A1 (en) | Automatic train tracking method and automatic train tracking system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20121219 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |