CN102509490A - Traffic information video acquisition experimental teaching method of modular design - Google Patents
Traffic information video acquisition experimental teaching method of modular design Download PDFInfo
- Publication number
- CN102509490A CN102509490A CN2011103092590A CN201110309259A CN102509490A CN 102509490 A CN102509490 A CN 102509490A CN 2011103092590 A CN2011103092590 A CN 2011103092590A CN 201110309259 A CN201110309259 A CN 201110309259A CN 102509490 A CN102509490 A CN 102509490A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- traffic
- video
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Closed-Circuit Television Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a traffic information video acquisition experimental teaching method of modular design. In the method, based on the acquired basic theory part, traffic parameter detection part and typical traffic violation judging part, the video processing step is made into multiple independent modules through programming; the acquisition of traffic volume is finished through the combined construction and parameter setting of a graying module, an image preprocessing module, a background extraction module and a virtual coil module; and a speed detection system is established by a dual-coil or single-coil method respectively to finish fixed-point speed acquisition and average speed calculation. A wrong-direction driving judging module and a violation parking judging module are established and connected respectively to realize the establishment of a vehicle wrong-direction driving and violation parking experiment system. In the invention, each step of the video processing is made into an independent module to finish an independent function. A student implements various video processing and application functions through reasonable matching and connection of the modules.
Description
Technical field
The invention belongs to transport information video acquisition teaching field; Based on transport information and control experiment porch; With the modular design is core, comprises the teaching and experiment method of the function such as basic theory, DETECTION OF TRAFFIC PARAMETERS, typical break in traffic rules and regulations differentiation of video acquisition, and builds teaching experiment system.
Background technology
Means of transportation are the important bases of national economy, are the advanced sectors of national economy.Be accompanied by informationization, intellectuality, the arrival in hommization epoch, the ITS technology is in the traffic and transport field ever more important.
Intelligent transportation system be many subjects fusion with intersect, its application promoted the raising of transport science and techonologies levels, played to improve conevying efficiency, reduce influence to environment, improve effects such as security and reliability.Therefore, be that the development of ITS of core will improve the quality of communications and transportation in all directions with using with the infotech.We can say that the transport information control technology is the basis of ITS development.In transport information and control technology; Video processing technique is occupied critical role; Should technology numerous areas such as traffic monitoring, volume of traffic detection, traffic safety and break in traffic rules and regulations detection be widely used at present, for the improvement and the obtaining of traffic data of traffic order provides effective technical means comprehensively.Because this technology has multidisciplinary intersection, has proposed new challenge for talents cultivating mode in recent years.How to make the student who is about to be engaged in traffic engineering field related work can grasp the gordian technique that transport information is handled? How on the basis of grasping information processing technology ultimate principle; Application-oriented aspect; Improve the ability of student's integrated application knowledge, just become the difficult point of classroom instruction, practical teaching, students ' quality cultivation.
At present; About correlated curriculum from transport information and control; Particularly find to the classroom instruction process of the relevant contents of courses such as IMAQ, Video processing; Because the characteristic of its subject crossing, there is certain problem in the aspects such as concrete structure of the difficulty that the teacher lectures on the classroom, the degree of depth and students practise link, mainly comprise:
(1), the static state demonstration mode of PPT courseware is mainly adopted in classroom instruction; Although a lot of teaching softwares is arranged to IMAQ and processing; But be that difficulty, the degree of depth all do not satisfy student's demand of traffic engineering specialty for the professional student of Flame Image Process offers basically.
(2), put into practice the mode that link is mainly taked visiting and learning, the student is in the understanding aspect for the application of classroom knowledge, the core technology of data acquisition and processing (DAP) is " black box " to the student basically, is difficult to reach the level of flexible Application.
(3) although image processing techniques has been successfully applied to traffic administration and applications such as the collection, break in traffic rules and regulations detection of transport information; But road conditions to urban transportation; The accuracy of this technology and high efficiency still exist some key issues to need to be resolved hurrily, and therefore need to carry out more deep research to various image processing methods.
To above 3 points, the high property of participation of needs of students one cover, high open teaching platform are with the cultivation demand of reply traffic engineering specialty for the student.For this reason, the present invention is around the video image acquisition treatment technology, and support Beijing University of Technology transport information and control laboratory platform have been developed the high property of participation of a cover, high opening, started to put into practice big teaching and experiment method in space and experimental system.
Summary of the invention
The objective of the invention is to; Through a kind of modular transport information video acquisition Experimental Teaching Method is provided; Development moduleization, cordwood traffic information collection teaching experiment system promote the transformation of traditional illustrative teaching pattern to property of participation, experience property teaching pattern for the student provides the platform of putting into practice about traffic video acquisition technique basic theory.
The present invention adopts following technological means to realize:
A kind of transport information video acquisition Experimental Teaching Method of modular design comprises may further comprise the steps basic theory, DETECTION OF TRAFFIC PARAMETERS, the typical case teaching experiment system of differentiating violating the regulations of video acquisition:
Basic theory part, DETECTION OF TRAFFIC PARAMETERS part, typical break in traffic rules and regulations judegment part to gather are divided into the basis, and the step of Video processing is passed through programming, are made into a plurality of independently modules; Each module has independently video processing function, has independently to import, export splicing ear, realizes the transmission of data stream through the linking between the terminal; Module is provided with scroll bar, the check box of realizing the Video processing parameter; Module comprises: gray processing module, image pre-processing module group, background extracting module, virtual coil module; Basic theory partly comprises: realize image gray processing, image binaryzation, the basic morphology operations of image, image advanced form mathematical operations, image smoothing and filtering operation; DETECTION OF TRAFFIC PARAMETERS partly realizes comprising: the volume of traffic detects, the speed of a motor vehicle detects, average speed detects; Typical case's break in traffic rules and regulations judegment part branch comprises: realize retrograde detection violating the regulations, the differentiation that red light violation detects, parking offense detects three kinds of break in traffic rules and regulations behaviors.
Combination through gray processing module, image pre-processing module group, background extracting module, virtual coil module is built and the parameter setting, accomplishes the collection of the volume of traffic; Use twin coil or unicoil method, build vehicle speed detection system respectively, accomplish the collection of the fixed point speed of a motor vehicle and the calculating of average speed.
Through gray processing module, image pre-processing module group, the combination of background extracting module, virtual coil module and signal lamp module is connected accomplishes building of red light violation recognition system; Use to drive in the wrong direction the building respectively and be connected of judge module and parking offense judge module realizes that vehicle drives in the wrong direction and the building of parking offense experimental system.
Aforesaid teaching experiment system is built with the traffic information collection hardware based on Video processing, through the traffic behavior on camera, the video frequency collection card capture platform, as the information source of Video processing.
Aforesaid Video processing comprises: decoding video stream, background initialization, vehicle characteristics extraction, vehicle tracking and context update.
Aforesaid decoding video stream is the video playback module based on video file, or the Online Video AM access module of using FFmpeg to separate scrambler based on experiment porch.
Aforesaid background is initialized as, the image background in the road environment except information of vehicles; Background image can not have vehicle image by a width of cloth and obtains, and can obtain through the computing to image sequence yet.
Aforesaid context update for background frames pixel value and current frame pixel value through weighted sum, obtain the pixel value of new background frames; Only need during renewal the pixel of background area is upgraded, the pixel point value of moving region then remains unchanged.
The present invention compared with prior art has following remarkable advantages and beneficial effect:
(1), the teaching scene is moved to the laboratory, the student can be participated in the learning process.Change the traditional teaching way of teacher's leading position, acceptance that the student is passive, make the student in practical operation and experiment are participated in, accomplish classroom learning and course practice of innovation.
(2), experimental system of the present invention adopts brick pattern, modular design mode; The function of each Flame Image Process all realizes with building through the reasonable link of disparate modules, makes the pilot process and the concrete function of the STUDENT ground grasp Flame Image Process of traffic engineering.What deserves to be mentioned is that functions such as different traffic information collections and detection violating the regulations can realize through the various combination of each module.This has fundamentally been avoided the conventional teaching method to cause producing the phenomenon of thinking set, and fully excites student's creativity consciousness.
(3), because experimental system adopts modularization, cordwood structure form, make system itself also become a experiment porch based on video acquisition, it has very strong exploitation expansion capability, and the space of independent research is provided for the student.The student can through using simple interface function, accomplish function module design and exploitation according to the experimental design demand of oneself, possesses the characteristic of autonomous experiment.Improved the student and independently studied ability, system itself has excellent generalization values and application prospect.
Description of drawings
Fig. 1 is volume of traffic testing process figure;
Fig. 2 is the testing process figure violating the regulations that drives in the wrong direction;
Fig. 3 is red light violation testing process figure;
Fig. 4 is parking offense testing process figure.
Embodiment
Explain below in conjunction with the Figure of description specific embodiments of the invention:
1, carry out modular design: modular design is meant that each step with Video processing is made into independently module through programming; Each module has independently video processing function; Have input, output splicing ear, realize the transmission of data stream through the rational linking between the terminal.Module has the parameter modification function, can realize the revision of parameter in the video processing procedure through operating units such as the scroll bar on the change module, check boxes.
Reasonable combination and linking through gray processing module, background extracting module, virtual coil module and red signal module can be accomplished building of red light violation recognition system.
In the performance history of whole experimental system, all embodied this design concept.The video processing procedure that relates in the basic step that video image is handled, the volume of traffic collection of handling based on video image, the processes such as detection violating the regulations based on Video processing; All development and Design becomes processing unit; Constitute " building block module ", the parameter of these modules can be carried out suitable adjustment to change the controlled variable in the image processing process through the operating unit on the module.In experimentation, the student can build different " building block module " voluntarily according to the experiment demand, realizes that the experimental design of oneself needs, and then carries out effect demonstration and evaluation intuitively through input, output module.
2, the traffic information collection teaching experiment system function design of handling based on video image:
From the teaching demand of traffic information collection technology, the present invention mainly comprises video image processing basic theory, DETECTION OF TRAFFIC PARAMETERS, three types of functions of typical break in traffic rules and regulations event detection.The realization difficulty of these three types of functions and complexity from the superficial to the deep meet the students'cognition process.
(1), the video image processing basis is theoretical
Video image processing basis theory part mainly comprises:
Gray processing: the gray processing computing formula of image is shown below
Y＝0.212671*R+0.715160*G+0.072169*B
Through three component substitutions of image RGB following formula just being obtained the gray-scale value of every of image.This operation can realize through the cvCvtColor function that calls among the OpenCV.
Binaryzation: the binaryzation of image realizes the cvThreshold function among the corresponding OpencV through the threshold operation to image.In order to embody the importance that threshold value is handled image binaryzation, in the system threshold value in the binary conversion treatment module is imported as the User Defined parameter with the form of slider bar, to embody the influence of threshold value to the binary conversion treatment result.
The basic morphology operations of image: the morphological operation of image comprises that to the erosion operation of image and dilation operation the erosion operation of image is accomplished by the cvErode function calculation, and the dilation operation of image is accomplished by the cvDilate function calculation.
Image advanced form mathematical operations: the image advanced form is learned opening operation, the closed operation of operation enclosed mass to image.In OpenCV, the computing of image advanced form is realized that by cvMorphologyEx the difference setting through to this function parameter can realize the advanced form mathematical operations.
Image smoothing with: image smoothing and filtering operation comprise computings such as medium filtering, mean filter, histogram equalization.
Filtering operation: in OpenCV, the image smoothing computing is realized that by cvSmoth the image histogram balancing operational is realized by cvEqualizeHist.
(2), DETECTION OF TRAFFIC PARAMETERS
The volume of traffic detects: the volume of traffic is meant the vehicle number that passes through a certain section of road in the unit interval, is called the magnitude of traffic flow or flow again.The volume of traffic has many types, wherein with the tool experiment meaning of hourly traffic volume and peak hour traffic.Experimental system is accomplished the detection to two kinds of volume of traffic on experiment porch.
Hourly traffic volume is meant the vehicle number that passes through a certain section of road in a hour.For hourly traffic volume, because one hour overlong time of actual measurement so when experiment, carry out detecting in 15 minutes, is promptly surveyed the 15min volume of traffic.The detection thinking is a counting method, in video, adds virtual coil, when vehicle crosses virtual coil, counts, and writes down in 15 minutes the vehicle number through coil, the promptly 15 minutes volume of traffic with this.
Peak hour traffic is meant a certain hour the highest volume of traffic of the volume of traffic in a day 24 hours.In experimental system, will shorten to 20 minutes the time, and detect wherein the volume of traffic the highest 2 minutes.Detect thinking on road, laying virtual coil and writing down in 20 minutes each car constantly through virtual coil; Detected the end back added up the volume of traffic in each continuous two minutes and selects mxm. at 20 minutes; Then the period at mxm. place be the peak " hour ", mxm. be the peak " hour " volume of traffic.Figure is as shown in Figure 1 for volume of traffic testing process.
The speed of a motor vehicle detects: the place speed of a motor vehicle (the instantaneous speed of a motor vehicle) is the main detected object during traffic video detects.The place speed of a motor vehicle is meant the speed of a motor vehicle when vehicle passes through a certain place of road (road section), also claims the instantaneous speed of a motor vehicle, and it is an important parameter of describing road somewhere point traffic.
When carrying out place speed of a motor vehicle detection, through one very short-range period, the ratio of this segment distance and time then is the place speed of a motor vehicle to experimental system with registration of vehicle.System will be provided with two virtual coils at the two ends in this section path and sail the time into and roll away from the time to confirm vehicle, thereby calculate travel speed on the path.
What deserves to be mentioned is because system adopts modular design, more than the detection method of two kinds of traffic parameters fixing, the student can set up the different detection flow process to achieve the goal through the different module collocation; The traffic parameter that software can detect also is not limited to above two kinds, and the selection through module and the collocation of module can also realize the detection of other traffic parameters, such as time headway, occupation rate etc.
(3), typical traffic events detects
Drive in the wrong direction and break rules and regulations to detect: the collocation through disparate modules has two kinds of detection modes violating the regulations of driving in the wrong direction.
Adopting twin coil to realize driving in the wrong direction detects
The virtual coil of one group of two close together is set on road, judges through the sequencing of two coils whether vehicle drives in the wrong direction through confirming vehicle.
Adopting vehicle tracking to realize driving in the wrong direction detects
Following the tracks of the thinking that detects is to have determined whether that at first vehicle sails detection zone into, then follows the tracks of the driving trace that this car is also confirmed vehicle if there is vehicle to sail into, judges through track whether vehicle drives in the wrong direction again.Implementation method is virtual coil to be set judged whether that vehicle sails detection zone on road, and utilization vehicle recognizer identification vehicle uses track algorithm to follow the tracks of vehicle then, judges whether the track of vehicle direction is retrograde violating the regulations, and it is as shown in Figure 2 to detect the differentiation process.
Red light violation detects: whether the thinking that red light violation detects is at first to judge whether to be red light phase, have vehicle to roll stop line away from if red light phase then detects, and then is judged as red light violation if having.Concrete implementation method judges whether to be red light phase for the teleseme signal lamp status signal through experiment porch at first, virtual coil is set roll away to have judged whether vehicle before the stop line of video image, thereby realizes that red light violation judges.It is as shown in Figure 3 to detect the differentiation process.
Parking offense detects: based on the hardware condition of experiment porch, during the crossing green light phase, the vehicle that does not pass through the crossing all is judged as parking offense in the experimental system.Whether the thinking that parking offense detects is at first to follow the tracks of vehicle, judges then whether vehicle moves, be green light phase according to the crossing, judges whether parking offense of vehicle.It is as shown in Figure 4 to detect the differentiation process.
The image processing function of native system is realized by OPENCV, and OPENCV is based on the VC++ platform development.In view of functional requirement, the present invention has selected to use the MFC engineering under the Visual C++.All modules are to belong to the class by the derivative CProcessing ModuleBase of CDialog class in the engineering, and the function of using in the module, variable etc. are encapsulated in this type.Through this type, can cut out the appearance of module at an easy rate and be operating units such as module interpolation scroll bar, check box.Because all modules all belong to identical base class, have so just accomplished the equality of module, the total setting of simultaneously a lot of intermodules can be unified in completion setting in the base class.
Separate module after the generation can't independent operating, and the work of most of module all depends on the output signal of higher level's module.For realizing output signal with higher level's module as this module input signal, and the result of this module is exported to subordinate's module, the both sides of each module all have an input terminal and a lead-out terminal at least, and this function realizes through Pin in program.Be connected with Pin_In and can two modules be coupled together through Pin_Out.Through this virtual terminal, data such as image, character string promptly can be in the intermodule transmission.Therefore,, will produce lasting data stream between interconnected module,, just can obtain result at last original video through the processing of modules at different levels to this data stream as long as source module continues to carry out data output.
The Data Matching verification scheme that The software adopted is following: program all can be judged the trigger terminal data type when each connecting moves takes place; Confirm to connect both sides' terminal data and whether be all triple channel image, single channel image or string data etc.; If two side's types are coincide; Then successful connection reports an error connection failure otherwise eject dialog box.
System design the cut-in method of two kinds of video informations, a kind of video playback module that is based on video file can insert any video file and carry out follow-up video image and handle.A kind of in addition Online Video AM access module that is based on experiment porch.For Online Video, system has used FFmpeg to separate scrambler, and it provides and has recorded, the total solution of conversion and fluidisation audio frequency and video.To the Online Video of transport information with the control experiment porch, system will change into image sequence as video source from the Online Video signal of camera.
The background initialization: the road background is meant the image background except information of vehicles in the road environment.Background image can be taken a width of cloth by manual work not to be had vehicle image and obtains, and can obtain through the computing to image sequence yet.The method of background extracting has a variety of, for example: the image time average, according to the frequency of occurrences in the image sequence differentiate, according to frequency of occurrences differentiation in the single image etc.In conjunction with experiment porch traffic background of the present invention characteristic of simple relatively, the main image time averaging method that adopts obtains background image in the system.
The proposition of image time average method is mainly according to the diversity of vehicle; The value on the ratio road surface that the brightness value of the pixel on the vehicle has is high, and the value on the ratio road surface that has is low, and vehicle has various colors; The vehicle major part of crossing all is kept in motion, and idol has of short duration parking to wait the vehicle of lamp.Therefore, the time " proportion " that background pixel point occurs is greater than the time of vehicle and other motion objects appearance, vehicle-state approximate random state.Therefore, the variation of the pixel point value of video image is near Gauss normal distribution.So from statistical angle, the variation that causes because of the process of each vehicle in the image can be ignored after average long-time.Thereby can utilize one section image sequence of vehicle operating to carry out the Background that time average obtains road.
The average of Gauss normal distribution (think here that continuous multiple frames image ask average average) is in theory near the actual value of background pixel.After the continuous hundreds of two field pictures to traffic scene added up to average, the image approximate that obtains was in real background.Be shown below,
Wherein (x y) representes reference background frame, F to B
k(x, y) the K frame in the presentation video sequence.
Vehicle characteristics extracts: obtain after the background image, it is poor current frame image and background image to be done, and obtains poor two field picture, suc as formula shown in:
D
t(x，y)＝|F
t(x，y)-B
t(x，y)|
To differ from two field picture and cut apart, the binary image that can obtain to move with threshold method.After the binaryzation in the image noise point more, utilize morphological method that binary image is handled again, like corrosion and dilation operation, finally obtain clear, do not have a binary image make an uproar.
Leaching process " trigger " is served as by virtual coil, passes through in case virtual coil detects vehicle, and coil can be given instruction of program, starts counting, extracting and trace routine to this vehicle.In addition, system also needs registration of vehicle to sail, leave time of virtual coil into, and temporal information is that the realization of the follow-up function such as test the speed of system provides foundation.
System takes following discriminant approach: when vehicle sails virtual coil into, a large amount of motion pixels can occur in the virtual coil, can it be regarded as " rising edge " of virtual coil state.And when vehicle left virtual coil, the motion pixel quantity in the virtual coil can reduce rapidly, can it be regarded as " negative edge " of virtual coil state.Through detecting the rising edge and the negative edge of virtual coil, can know clearly that vehicle sails and leave the time of virtual coil into.Simultaneously, system adopts " human eye recognition feature " to obtain the concrete characteristic information of vehicle.When vehicle passes through virtual coil, be equivalent to vehicle and get into the human eye identified region; Area-of-interest and the difference image between the background image that system's utilization is in the virtual coil top grasp the vehicle characteristics piece, are equivalent in the human eye identified region, find the concern characteristic of vehicle.
Vehicle tracking: vehicle triggers the vehicle characteristics block search through behind the virtual coil.System obtains characteristic block, begins from next frame, and be unit area (20 * 20) with the area of this target square, travel through the whole field of search, seek the piece that matees most with this characteristic block.The field of search is the coordinate center with the characteristic block, is set at nine times of sizes (60 * 60) of characteristic block.
Traversal search begins from the upper left corner, the field of search, grasps with size such as this target, and as to be detected, itself and characteristic block are done difference can be got:
R＝|R1-R2|
In the formula, the average red component of R1 representation feature piece, R2 are represented to be detected average red component, and R is both differences.The difference computing method of other two average color components are identical therewith.
Calculate the difference and the preservation of three average color components and characteristic block respectively; The coordinate pixel that moves to right grasps image block more then, again with characteristic block do poor; If three-component poor sum is less than the lowest difference of preserving in the past; Then write down this square coordinate, and upgrade the difference of three average color components, otherwise do not do any operation.Coordinate moves a pixel once more, and behind the right margin of the arrival field of search, ordinate adds one, continues from left to right traversal, travels through whole region of search with this.Final preserve be exactly in the whole region of search with the just the most close piece of characteristic block difference minimum, the record coordinate, every frame is done same operation, realizes the tracking to the vehicle characteristics piece
Context update: experiment scene is to continue to change, and this carries out real-time update to background after needing each two field picture to finish dealing with.To certain point in the background, do weighted sum with background frames pixel value and current frame pixel value, obtain the pixel value of new background frames.Only need during renewal the pixel of background area is upgraded, and the pixel point value of moving region will remain unchanged.The present frame weights play an important role to background extracting, and less weights reduce vehicle to the disturbance of background but context update speed is slow, and bigger weights can reflect current background in real time, but receive vehicle interference itself bigger.
System specifically uses step very simple; Only need pulling and be connected through each module in the system; Concrete experiment according to the user need be built experimentation; And utilizing the visual output result and the variation characteristics of image in input, the output module, the volume of traffic of let ultimate principle that the students understand video image handles and step, handling based on video image obtains technology, based on the break in traffic rules and regulations incident discrimination technology of video image processing.
Teaching experiment system of the present invention provides abundant experiment module, amounts to 33 kinds.And with the functional character of color differentiating module.Wherein pale yellow representative image processing-image transformation class, brown representative image processing-filtering class, sky blue representative image processing-morphological image is handled class; The green advanced processes class of representing, the pink colour representative data stores class, grass green representative data conversion class; On behalf of conventional data, purple show class; Red representative image processing-Tu image source class, purplish blue look representative data transmitting-receiving class, grey representative image processing-data statistics class.Each functions of modules is shown in table one:
Table one
Based on the separate module of these Flame Image Process, any reasonable combination of intermodule will be accomplished the ultimate principle experiment of Flame Image Process and based on the transport information application experiment of video image processing technology.
Claims (6)
1. the transport information video acquisition Experimental Teaching Method of a modular design comprises it is characterized in that basic theory, DETECTION OF TRAFFIC PARAMETERS, the typical case teaching experiment system of differentiating violating the regulations of video acquisition, may further comprise the steps:
1.1. basic theory part, DETECTION OF TRAFFIC PARAMETERS part, typical break in traffic rules and regulations judegment part to gather are divided into the basis, and the step of Video processing is passed through programming, are made into a plurality of independently modules; Each module has independently video processing function, has independently to import, export splicing ear, realizes the transmission of data stream through the linking between the terminal; Module is provided with scroll bar, the check box of realizing the Video processing parameter;
Described module comprises: gray processing module, image pre-processing module group, background extracting module, virtual coil module;
Described basic theory partly comprises: realize image gray processing, image binaryzation, the basic morphology operations of image, image advanced form mathematical operations, image smoothing and filtering operation;
Described DETECTION OF TRAFFIC PARAMETERS partly realizes comprising: the volume of traffic detects, the speed of a motor vehicle detects, average speed detects;
Described typical break in traffic rules and regulations judegment part branch comprises: realize retrograde detection violating the regulations, the differentiation that red light violation detects, parking offense detects three kinds of break in traffic rules and regulations behaviors;
1.2. the combination through gray processing module, image pre-processing module group, background extracting module, virtual coil module is built and the parameter setting, accomplishes the collection of the volume of traffic; Use twin coil or unicoil method, build vehicle speed detection system respectively, accomplish the collection of the fixed point speed of a motor vehicle and the calculating of average speed;
1.3. through gray processing module, image pre-processing module group, the combination of background extracting module, virtual coil module and signal lamp module is connected accomplishes building of red light violation recognition system; Use to drive in the wrong direction the building respectively and be connected of judge module and parking offense judge module realizes that vehicle drives in the wrong direction and the building of parking offense experimental system.
2. the transport information video acquisition Experimental Teaching Method of modular design according to claim 1; It is characterized in that: described teaching experiment system; Be built with traffic information collection hardware based on Video processing; Through the traffic behavior on camera, the video frequency collection card capture platform, as the information source of Video processing.
3. the transport information video acquisition Experimental Teaching Method of modular design according to claim 1 is characterized in that described Video processing comprises: decoding video stream, background initialization, vehicle characteristics extraction, vehicle tracking and context update.
4. the transport information video acquisition Experimental Teaching Method of modular design according to claim 3; It is characterized in that: described decoding video stream is for based on the video playback module of video file, or the Online Video AM access module of using FFmpeg to separate scrambler based on experiment porch.
5. the transport information video acquisition Experimental Teaching Method of modular design according to claim 3, it is characterized in that: described background is initialized as, the image background in the road environment except information of vehicles; Background image can not have vehicle image by a width of cloth and obtains, and can obtain through the computing to image sequence yet.
6. the transport information video acquisition Experimental Teaching Method of modular design according to claim 3 is characterized in that: described context update for background frames pixel value and current frame pixel value through weighted sum, obtain the pixel value of new background frames; Only need during renewal the pixel of background area is upgraded, the pixel point value of moving region then remains unchanged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103092590A CN102509490A (en) | 2011-10-13 | 2011-10-13 | Traffic information video acquisition experimental teaching method of modular design |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103092590A CN102509490A (en) | 2011-10-13 | 2011-10-13 | Traffic information video acquisition experimental teaching method of modular design |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102509490A true CN102509490A (en) | 2012-06-20 |
Family
ID=46221566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011103092590A Pending CN102509490A (en) | 2011-10-13 | 2011-10-13 | Traffic information video acquisition experimental teaching method of modular design |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102509490A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105518759A (en) * | 2013-09-06 | 2016-04-20 | 罗伯特·博世有限公司 | Method and control and detection device for checking the plausibility of a motor vehicle driving in the wrong direction |
| CN106199133A (en) * | 2016-06-27 | 2016-12-07 | 张超超 | Current sample processing platform |
| CN108428347A (en) * | 2018-05-14 | 2018-08-21 | 三峡大学 | A kind of vehicle count method and system based on more detection zones |
| CN109712428A (en) * | 2019-02-27 | 2019-05-03 | 爱笔（北京）智能科技有限公司 | A kind of parking position localization method and device, system |
| CN111127877A (en) * | 2019-11-19 | 2020-05-08 | 华为技术有限公司 | Road condition information monitoring method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101510358A (en) * | 2009-03-20 | 2009-08-19 | 吉林大学 | Method and apparatus for processing real time statistical vehicle flowrate using video image |
| US20090207046A1 (en) * | 2006-03-22 | 2009-08-20 | Kria S.R.L. | system for detecting vehicles |
| CN101593422A (en) * | 2009-06-15 | 2009-12-02 | 无锡骏聿科技有限公司 | A kind of traffic violation detection method based on video frequency tracking and pattern-recognition |
| CN101635852A (en) * | 2009-08-26 | 2010-01-27 | 北京航空航天大学 | Method for detecting real-time moving object based on adaptive background modeling |
-
2011
- 2011-10-13 CN CN2011103092590A patent/CN102509490A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090207046A1 (en) * | 2006-03-22 | 2009-08-20 | Kria S.R.L. | system for detecting vehicles |
| CN101510358A (en) * | 2009-03-20 | 2009-08-19 | 吉林大学 | Method and apparatus for processing real time statistical vehicle flowrate using video image |
| CN101593422A (en) * | 2009-06-15 | 2009-12-02 | 无锡骏聿科技有限公司 | A kind of traffic violation detection method based on video frequency tracking and pattern-recognition |
| CN101635852A (en) * | 2009-08-26 | 2010-01-27 | 北京航空航天大学 | Method for detecting real-time moving object based on adaptive background modeling |
Non-Patent Citations (3)
| Title |
|---|
| 刘肃亮: "交通车辆违章智能视频监控***的设计与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》, no. 200404, 30 April 2004 (2004-04-30) * |
| 张国强: "交通参数视觉实时检测与识别方法研究", 《中国优秀硕士学位论文全文数据库工程科技II辑》, no. 200508, 31 August 2005 (2005-08-31) * |
| 沈连丰: "通信新技术及其实验", 《通信新技术及其实验》, 30 September 2003 (2003-09-30) * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105518759A (en) * | 2013-09-06 | 2016-04-20 | 罗伯特·博世有限公司 | Method and control and detection device for checking the plausibility of a motor vehicle driving in the wrong direction |
| CN106199133A (en) * | 2016-06-27 | 2016-12-07 | 张超超 | Current sample processing platform |
| CN108428347A (en) * | 2018-05-14 | 2018-08-21 | 三峡大学 | A kind of vehicle count method and system based on more detection zones |
| CN109712428A (en) * | 2019-02-27 | 2019-05-03 | 爱笔（北京）智能科技有限公司 | A kind of parking position localization method and device, system |
| CN111127877A (en) * | 2019-11-19 | 2020-05-08 | 华为技术有限公司 | Road condition information monitoring method and device |
| WO2021098211A1 (en) * | 2019-11-19 | 2021-05-27 | 华为技术有限公司 | Road condition information monitoring method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108830150B (en) | One kind being based on 3 D human body Attitude estimation method and device | |
| CN107844635B (en) | System for realizing BIM information and traffic simulation information integration and integration method thereof | |
| CN102509490A (en) | Traffic information video acquisition experimental teaching method of modular design | |
| CN109800689A (en) | A kind of method for tracking target based on space-time characteristic fusion study | |
| CN106530890A (en) | Intelligent driving training system and method based on vehicle posture and trajectory perception | |
| CN109635853A (en) | The method for automatically generating artificial intelligence training sample based on computer graphics techniques | |
| CN111208842B (en) | Virtual unmanned aerial vehicle and entity unmanned aerial vehicle mixed cluster task control system | |
| CN110188683A (en) | A kind of automatic Pilot control method based on CNN-LSTM | |
| CN106228109A (en) | A kind of action identification method based on skeleton motion track | |
| CN104966320B (en) | A kind of pattern-painting automatic generation method based on three rank Bezier curves | |
| CN104408760A (en) | Binocular-vision-based high-precision virtual assembling system algorithm | |
| CN106781810A (en) | Wind-driven generator centering attended operation method based on 3D dummy emulation systems | |
| CN107067858A (en) | A kind of three-dimensional simulation training system of goat | |
| CN111259950B (en) | Method for training YOLO neural network based on 3D model | |
| CN105184229A (en) | Online learning based real-time pedestrian detection method in dynamic scene | |
| CN104200422A (en) | Expert system for remote sensing image processing | |
| CN109934106A (en) | A kind of user behavior analysis method based on video image deep learning | |
| CN109508686A (en) | A kind of Human bodys' response method based on the study of stratification proper subspace | |
| CN108791302A (en) | Driving behavior modeling | |
| CN116109455A (en) | Language teaching auxiliary system based on artificial intelligence | |
| CN103456012B (en) | Based on visual human hand detecting and tracking method and the system of maximum stable area of curvature | |
| CN108717522A (en) | A kind of human body target tracking method based on deep learning and correlation filtering | |
| CN106846459A (en) | A kind of collision checking method for VR Virtual Campus Systems | |
| CN103020984B (en) | Based on the 3 d human motion data complementing method of sparse expression | |
| Bai et al. | Analysis of classroom teacher-student interaction based on Yolo |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120620 |