CN105761504A - Vehicle speed real-time measuring method based on inhomogeneous video image frame collection - Google Patents
Vehicle speed real-time measuring method based on inhomogeneous video image frame collection Download PDFInfo
- Publication number
- CN105761504A CN105761504A CN201610327118.4A CN201610327118A CN105761504A CN 105761504 A CN105761504 A CN 105761504A CN 201610327118 A CN201610327118 A CN 201610327118A CN 105761504 A CN105761504 A CN 105761504A
- Authority
- CN
- China
- Prior art keywords
- image
- speed
- mask
- vehicle
- time
- 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
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Abstract
The invention provides a vehicle-speed real-time measuring method based on inhomogeneous video image frame collection and relates to the field of vehicle speed measuring methods. The speed is measured through the following steps that firstly, a frame of image is obtained from a camera and mapped into a system standard dimension, a mask of a lane to be detected is marked, an interested area based on the mask is generated, a speed measuring starting area mask part and a speed measuring ending area mask part are marked in the mask, and interested areas are generated respectively; image frames are obtained from the camera in sequence after practical calibration, and whether a vehicle arrives or not is judged according to the difference value of detection image starting windows of the interested area of the starting mask part; whether a vehicle arrives or not is judged according to the difference value of detection image ending windows of the interested area of the ending mask part; the vehicle speed is worked out according to the practical distance and the timing time of a system. According to the method, the camera and other front-end equipment can be directly used for carrying out speed measuring, the real-time performance is good, precision is high, and the anti-jamming capability is high.
Description
Technical field
The invention belongs to vehicle speed measuring method field, more specifically to a kind of vehicle speed measuring method based on video.
Background technology
Along with the development rapidly of society, in field of traffic, how to measure motor vehicles travel speed accurately and be always up the study hotspot of national science and technology workers.Video frequency speed-measuring, by obtaining the video information that camera collection arrives, extracts video background and target vehicle and by actual demarcation in advance by the method for software analysis, measures the speed of vehicle.
It is simple that video frequency speed-measuring has installation, it is not necessary to the plurality of advantages such as special speed measuring equipment, but there is also many problems needing to solve: is first that the precision of video frequency speed-measuring is as the change of light conditions and changes；Next to that shadow problem in video frequency speed-measuring algorithm, easily causing flase drop false retrieval, accuracy is not high；Finally, the vehicle problem mutually blocked in road scene.Current difficult point also focuses in the segmentation of vehicle, is considered as arranging multiple auxiliary detection region at algorithm design aspect, carries out multiple analytical calculation.It addition, the algorithm in present most of video frequency speed-measuring methods is complex, it is impossible to well requirement of real time, hardware must adopt the chip that processes more at a high speed meet the demand of advanced algorithm.
Summary of the invention
It is an object of the invention to propose a kind of video frequency speed-measuring method to solve that existing video frequency speed-measuring method poor real, precision be low, the problem such as poor anti jamming capability and other speed-measuring methods test equipment cost height, system complex.
Based on above target, the present invention proposes a kind of speed method for real-time measurement gathered based on non-homogeneous video frame image, and it comprises the steps:
1. system initialization: first obtain a two field picture from photographic head and be mapped as system standard size, this method suggestion is of a size of 1024 × 768；Then delimit the mask in track at suitable speed measuring position according to the shape in track to be measured；Secondly, automatically generate its minimum external square area-of-interest as this mask according to this mask, it addition, delimit test the speed initiation region mask and end region mask in this mask, and generate corresponding area-of-interest in the same way；Finally, extract initial, to terminate area-of-interest background, and carry out gray proces.Ensure test the speed accuracy premise under, use sense region-of-interest carries out speed measurement treatment.
2. actual demarcate: system according to test the speed initial, terminate mask and obtain its central point, measure actual ranges the input system of these 2 correspondences in track according to the two central point.
3. car speed calculates: obtains picture frame successively from photographic head, each two field picture is all handled as follows: update the data of all area-of-interests.Area-of-interest according to sintering mask detects the image difference of every two field picture start window and determines whether that car arrives；When start window has detected that car arrives, system carries out timing according to system clock；Constantly obtaining picture frame from photographic head, the image difference terminating window according to the area-of-interest every two field picture of detection terminating mask determines whether that car arrives；System is when processing each frame, if there being car to arrive, system timing terminates, and returns and tests the speed the time, goes out speed according to actual range and the Time Calculation that tests the speed and completes the function such as vehicle flowrate and overspeed detection.If arriving without car, system obtains next frame immediately and detects.
1., for the image difference detection in step 3. middle area-of-interest, mainly have the following steps:
(1) current interest region carries out gray proces；
(2) current interest region and background carry out calculus of differences under the auxiliary of mask and process acquisition difference equalization image: first, current interest area image and background image subtraction obtain error image g (x, y)=f (x, y)-h (x, y)；Then, find out pixel minimum M in error image, and allow each pixel of error image obtain the image being modified plus this numerical value；Finally, find out and be modified in image pixel maximum Max, and allow each pixel of the image being modified be multiplied by 255/Max, obtain final difference equilibrium figures as dst (x, y).
(3) image is carried out the binary conversion treatment of adaptive threshold: adopt big law to calculate threshold value on difference equilibrium figures picture, carry out binary conversion treatment according to threshold value and obtain binary image.
(4) add up the pixel number that gray value is 255 and account for the ratio of the total pixel of this area-of-interest；
(5) judge that whether ratio is less than the context update threshold value set, and if so, then updates the background in current interest region, otherwise, then terminate, be determined further.
2. when carrying out system initialization, having delimited three kinds of masks, and generated corresponding area-of-interest based on each mask, this area-of-interest is the minimum enclosed rectangle of mask；Ensure test the speed accuracy premise under, use sense region-of-interest carry out speed measurement treatment greatly reduce system process data volume, frame per second is improved, and therefore improves system real time and accuracy.
3. being based on, in the step object that 3. middle data process, the area-of-interest initial, end mask that tests the speed, and judging whether vehicle arrives the basis of start window is the mask in area-of-interest, system rejection to disturbance ability improves.
4. step 3. in difference detection prospect and background subtracting time use data be the valid data within area-of-interest and not all background data.
5. step 3. in determine whether vehicle arrive start window time be separately provided a flag bit HSTART, detect that when image difference is bigger, HSTART adds 1, otherwise, HSTART zero setting, can be judged to as HSTART > 1 vehicle arrive.Really there is vehicle to arrive so only just can regard as when monitoring twice difference continuously and being bigger, reduce the impact of part interference, enhance the robustness of system.
6. there is car to start timing when arriving start window and to the detection terminating window in step 3. middle judgement.
7. use system clock in the time of step 3. middle system timing, detect that car detects car to terminating window from start window, the time that the time of vehicle movement and this section of distance of system process, compared to by frame per second and the frame number long-pending time obtained, this system time difference is used to represent that the time of vehicle actual motion reduces systematic error, the time calculated is made to be as closely as possible to the time of vehicle actual motion, vehicle movement time precision is made to arrive Microsecond grade, and the time precision calculated by fixing frame per second is Millisecond, therefore improve the accuracy of time.
8. the vehicle foundation through the end window flag bit HEND as being separately provided is judged when step is 3. middle, terminate windows detecting to image difference bigger time this position can be made to add 1, otherwise, system being made when HEND is more than 1 to be judged to end window and detected that car arrives, timing terminates.
9. also include in the step 3. middle window place that terminates: if do not have vehicle to arrive, should abandon processing present frame then jump into the process to next frame immediately；Photographic head obtains video flowing, in the algorithm, system reads next frame immediately and processes after having processed a frame, so, reading, relative to timing, the method that frame of video processes, the frame of video that the method processes is relatively many, can obtain the positional information of more vehicle, space length reduces error, improves the accuracy of detection of speed.
The speed method for real-time measurement that gathers based on non-homogeneous video frame image involved in the present invention is the high-precision real hourly velocity detection that the method such as combining image difference detection realizes road vehicle on the basis of Computer Vision, has the advantage that
1. real-time is high, and major measure is embodied in following several respects:
(1) in method, data process is concentrated mainly on initial, end area-of-interest and image is carried out gray processing process, decreases the data volume that system processes；
(2) better simply algorithm is used when ensureing to test the speed effect so that system executive time decreased, it is possible to reflect vehicle actual motion situation better in real time.
2. the velocity accuracy calculated is high, and major measure is embodied in following several respects:
(1) time precision is high: the time of day of vehicle movement time is the order of magnitude is the system clock beat of more than Microsecond grade, but not the time that the order of magnitude is Millisecond calculated by frame per second；
(2) range accuracy is high: system can read next frame after having processed a frame immediately and carry out processing without wait, obtain the non-homogeneous video frame image that frame per second is higher, and system data treating capacity reduces, make system can obtain more vehicle position information, system is more accurate to the judgement of vehicle condition, thus decreasing systematic error on space length；
(3), during detection vehicle, use error image equalization, reduce the situation of vehicle false retrieval missing inspection.
3. capacity of resisting disturbance is strong, and major measure is embodied in following several respects:
(1) just can regard as when being consecutively detected twice difference and being bigger really have vehicle arrive detection zone；
(2) with initial, the end area-of-interest detection to track situation, and each area-of-interest is when carrying out data and processing, and all employ the track mask of correspondence, decreases the extraneous data impact on vehicle detection.
Accompanying drawing explanation
Fig. 1 is the main-process stream schematic diagram of the present invention
Fig. 2 is that present system initializes schematic diagram
Fig. 3 is image difference of the present invention detection schematic diagram
Fig. 4 is specific embodiment of the invention case effect figure video flow graph to be measured
Fig. 5 is specific embodiment of the invention case effect figure track to be measured, start window, end window mask delimitation figure
Fig. 6 is the mask figure in the area-of-interest of specific embodiment of the invention case effect figure track to be measured
Fig. 7 is that specific embodiment of the invention case effect figure detects that vehicle arrives the figure of start window situation
Fig. 8 is that specific embodiment of the invention case effect figure detects that vehicle arrives the figure terminating window situation
Fig. 9 is specific embodiment of the invention case effect figure target vehicle velocity display figure
Figure 10 is the simplification general flow chart of the present invention
Detailed description of the invention
In order to technical scheme is explained further, below by specific embodiment, the present invention will be described in detail.
As it is shown in figure 1, the particular flow sheet of a kind of vehicle speed measuring method core algorithm based on video for the present invention relates to.It is broadly divided into system initialization, actual demarcation, car speed three steps of calculating.
1. system initialization: concrete initialization flow process is as shown in Figure 2.
2. actual demarcation: before reality is demarcated, it is necessary to draw and test the speed initial, end window and return to its center, realize the demarcation of actual point at the scene according to the particular location of central point, and obtain actual range.
3. speed calculates: in testing the speed, and the flow process detected for the difference of image is as shown in Figure 3.
As shown in Fig. 4,5,6,7,8,9, particular flow sheet when it is implemented for present pre-ferred embodiments is actual, above-mentioned preferred steps is all merged wherein by it.
In the diagram, expression is the original size of video flowing to be measured.
In Figure 5, what represent is the track to be measured to video flowing to be measured, start window, the delimitation of end window mask, and system automatically generates track area-of-interest corresponding to mask, the start window that tests the speed area-of-interest, test the speed end window area-of-interest, system according to test the speed initial, terminate mask return its central point, according to actual calibration result, by these 2 actual range input systems, and initial, end window being carried out preliminary background extracting gray processing, idiographic flow refers to Fig. 2.
In figure 6, expression is the track to be measured mask in its area-of-interest.The use of area-of-interest, can largely reduce data processing amount, and the use of mask, more interference factor can be filtered.
In the figure 7, expression is the situation that system processes when there being vehicle to arrive track start window to be measured.When start window has detected that car arrives, this car of system acquisition, system starts timing, constantly end window is detected simultaneously.
In fig. 8, expression is when there being vehicle to arrive the situation that when window is terminated in track to be measured, system processes.When terminating window and having detected that car arrives, this car of system acquisition, system finishing timing, calculate target vehicle after out-of-date speed, start again constantly start window to be detected, detect the speed of next target vehicle.
In fig .9, expression is target vehicle velocity display situation.This figure is shown that track to be measured has had two cars through out-of-date speed size, and the time of display speed is to have car namely to show speed after detection zone.Here the hypervelocity threshold value used that tests the speed is 30km/h.
The detailed description of the invention of whole system is substantially and carries out according to the order of Fig. 4,5,6,7,8,9, and some of data are specifically processed mainly concrete steps according to the flow chart in Fig. 1,2,3 by system.Mode embodied herein is on computers the reasonable of algorithm to be verified with correctness, and program is based on photographic head front end data and processes and design, if the embedded data being specifically applied to photographic head front end processes, only need to this program be modified slightly.
Claims (7)
1. the speed method for real-time measurement gathered based on non-homogeneous video frame image, it is characterised in that comprise the steps:
1. system initialization: obtain a two field picture from photographic head and be mapped as system standard size；Delimit track to be detected mask and the area-of-interest based on this mask；In mask, delimit initial, the end zone mask that tests the speed, generate the area-of-interest of correspondence respectively；Extract initial, to terminate area-of-interest background, carry out gray proces.
2. actual demarcate: system according to test the speed initial, terminate mask and obtain its central point, measure actual ranges the input system of these 2 correspondences in track according to the two central point.
3. car speed calculates: obtaining picture frame successively from photographic head, the image difference detecting every two field picture start window according to the area-of-interest of sintering mask determines whether that car arrives；When start window has detected that car arrives, system carries out timing according to system clock；Constantly obtaining picture frame from photographic head, the image difference terminating window according to the area-of-interest every two field picture of detection terminating mask determines whether that car arrives；System is when processing each frame, if there being car to arrive, system timing terminates, and returns and tests the speed the time, goes out speed according to actual range and the Time Calculation that tests the speed.If arriving without car, system obtains next frame immediately and detects.
2. a kind of speed method for real-time measurement gathered based on non-homogeneous video frame image as claimed in claim 1, it is characterised in that ensure that accurately under premise, data volume processes less, and anti-interference is stronger:
1. step 1-1. in carry out system initialization time delimit three masks, and the area-of-interest of correspondence is generated based on each mask, ensure test the speed accuracy premise under, use sense region-of-interest carry out speed measurement treatment greatly reduce system process data volume, improve system real time.
2. step 1-3. in the detection of vehicle used the difference detection method of area-of-interest, the method is after the window that tests the speed of present image is carried out gray proces, difference equalization image (data outside mask are set to zero) is obtained with its background subtracting in mask, then medium filtering is carried out, adaptive threshold is adopted to carry out binary conversion treatment, the quantity calculating the pixel that gray value is 255 accounts for the ratio of total pixel quantity, this ratio is compared with the proportion threshold value of setting, show more than threshold value table and occur in that vehicle, represent there is no vehicle less than threshold value.This detection method is simple and efficiently, it is to avoid use complicated optical flow method, Corner Detection and mate, spot detection with the method such as mate, reduction data processing amount, raising real-time.
3. a kind of speed method for real-time measurement gathered based on non-homogeneous video frame image as claimed in claim 1, it is characterized in that, system clock is used: detect that car is to terminating the time that the windows detecting system time difference to car is vehicle movement from start window in the time of step 1-3. middle system timing, this system time difference is used to replace the time of vehicle actual motion, compare the time calculated by fixing frame per second, improve several order of magnitude, greatly reduce time error so that the time calculated is as closely as possible to the time of vehicle actual motion.
4. a kind of speed method for real-time measurement gathered based on non-homogeneous video frame image as claimed in claim 1, it is characterized in that, system adopt non-homogeneous video frame image acquisition method: step 1-3. in, if do not have vehicle to arrive end window, abandon immediately processing present frame and obtaining next frame image from photographic head processing, middle without waiting for timer time.The image obtained is non-homogeneous video frame image, and frame per second is of a relatively high, and vehicle actual motion process sampling number is more so that system obtains the positional information of more vehicle, reduces systematic error on space length.
5. the speed method for real-time measurement that a kind of non-homogeneous video frame image as claimed in claim 1 gathers, it is characterized in that, to vehicle at speed trial ground from initiateing detection of end and to calculate the process of speed practical: step 1-3. determine whether vehicle arrive initial, terminate window time be respectively provided with a flag bit, each detect that when image difference is bigger, respective flag position adds 1, this position zero when image difference is less, when this flag bit is more than 1, namely detected twice after satisfying condition, vehicle can be judged to and arrived.So only when double detecting just can determine whether when image difference is bigger that vehicle arrives start window or terminates window, reducing the impact of part interference, enhancing the robustness of system.
6. a kind of speed method for real-time measurement gathered based on non-homogeneous video frame image as claimed in claim 1, it is characterized in that, it is based on the area-of-interest initial, end mask that tests the speed in the step 1-object that 3. middle data process, and judging whether vehicle arrives the basis of start window is the mask in area-of-interest, not conventional two-coil method.
7. a kind of speed method for real-time measurement gathered based on non-homogeneous video frame image as claimed in claim 1, it is characterized in that, for in the image difference detection of step 1-3. middle area-of-interest, employ error image equalization method: current interest region and background carry out calculus of differences under the auxiliary of mask and process acquisition difference equalization image: first, current interest area image and background image subtraction obtain error image g (x, y)=f (x, y)-h (x, y)；Then, find out pixel minimum M in error image, and allow each pixel of error image obtain the image being modified plus this numerical value；Finally, find out and be modified in image pixel maximum Max, and allow each pixel of the image being modified be multiplied by 255/Max, obtain final difference equilibrium figures as dst (x, y).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610327118.4A CN105761504B (en) | 2016-05-17 | 2016-05-17 | Speed method for real-time measurement based on the collection of non-homogeneous video frame image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610327118.4A CN105761504B (en) | 2016-05-17 | 2016-05-17 | Speed method for real-time measurement based on the collection of non-homogeneous video frame image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105761504A true CN105761504A (en) | 2016-07-13 |
| CN105761504B CN105761504B (en) | 2018-02-09 |
Family
ID=56324133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610327118.4A Expired - Fee Related CN105761504B (en) | 2016-05-17 | 2016-05-17 | Speed method for real-time measurement based on the collection of non-homogeneous video frame image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105761504B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315095A (en) * | 2017-06-19 | 2017-11-03 | 哈尔滨工业大学 | Many vehicle automatic speed-measuring methods with illumination adaptability based on Video processing |
| CN113506449A (en) * | 2021-05-21 | 2021-10-15 | 郑州大学 | High-speed highway vehicle speed measuring method based on video compression domain |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6795567B1 (en) * | 1999-09-16 | 2004-09-21 | Hewlett-Packard Development Company, L.P. | Method for efficiently tracking object models in video sequences via dynamic ordering of features |
| US20050114103A1 (en) * | 2003-09-09 | 2005-05-26 | Bohyung Han | System and method for sequential kernel density approximation through mode propagation |
| CN101141633A (en) * | 2007-08-28 | 2008-03-12 | 湖南大学 | Moving object detecting and tracing method in complex scene |
| CN101303701A (en) * | 2008-06-27 | 2008-11-12 | 北京中星微电子有限公司 | Method and apparatus for searching video |
| CN101615342A (en) * | 2008-06-27 | 2009-12-30 | 青岛海信电子产业控股股份有限公司 | A kind of vehicle checking method |
| CN102136196A (en) * | 2011-03-10 | 2011-07-27 | 北京大学深圳研究生院 | Vehicle velocity measurement method based on image characteristics |
| CN102768731A (en) * | 2012-06-29 | 2012-11-07 | 陕西省交通规划设计研究院 | Method and system for automatic positioning and identifying target based on high definition video images |
| CN102810203A (en) * | 2012-06-29 | 2012-12-05 | 陕西省交通规划设计研究院 | System and method for detecting unknown article and event based on high-definition video monitoring image |
| CN102902951A (en) * | 2012-06-29 | 2013-01-30 | 陕西省交通规划设计研究院 | System and method for vehicle target location and event detection on basis of high-definition video monitoring images |
-
2016
- 2016-05-17 CN CN201610327118.4A patent/CN105761504B/en not_active Expired - Fee Related
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6795567B1 (en) * | 1999-09-16 | 2004-09-21 | Hewlett-Packard Development Company, L.P. | Method for efficiently tracking object models in video sequences via dynamic ordering of features |
| US20050114103A1 (en) * | 2003-09-09 | 2005-05-26 | Bohyung Han | System and method for sequential kernel density approximation through mode propagation |
| CN101141633A (en) * | 2007-08-28 | 2008-03-12 | 湖南大学 | Moving object detecting and tracing method in complex scene |
| CN101303701A (en) * | 2008-06-27 | 2008-11-12 | 北京中星微电子有限公司 | Method and apparatus for searching video |
| CN101615342A (en) * | 2008-06-27 | 2009-12-30 | 青岛海信电子产业控股股份有限公司 | A kind of vehicle checking method |
| CN102136196A (en) * | 2011-03-10 | 2011-07-27 | 北京大学深圳研究生院 | Vehicle velocity measurement method based on image characteristics |
| CN102768731A (en) * | 2012-06-29 | 2012-11-07 | 陕西省交通规划设计研究院 | Method and system for automatic positioning and identifying target based on high definition video images |
| CN102810203A (en) * | 2012-06-29 | 2012-12-05 | 陕西省交通规划设计研究院 | System and method for detecting unknown article and event based on high-definition video monitoring image |
| CN102902951A (en) * | 2012-06-29 | 2013-01-30 | 陕西省交通规划设计研究院 | System and method for vehicle target location and event detection on basis of high-definition video monitoring images |
Non-Patent Citations (1)
| Title |
|---|
| 董春利 等: ""基于视频的车辆检测与跟踪算法综述"", 《南京邮电大学学报（自然科学版）》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315095A (en) * | 2017-06-19 | 2017-11-03 | 哈尔滨工业大学 | Many vehicle automatic speed-measuring methods with illumination adaptability based on Video processing |
| CN107315095B (en) * | 2017-06-19 | 2019-07-02 | 哈尔滨工业大学 | More vehicle automatic speed-measuring methods with illumination adaptability based on video processing |
| CN113506449A (en) * | 2021-05-21 | 2021-10-15 | 郑州大学 | High-speed highway vehicle speed measuring method based on video compression domain |
| CN113506449B (en) * | 2021-05-21 | 2022-06-21 | 郑州大学 | High-speed highway vehicle speed measuring method based on video compression domain |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105761504B (en) | 2018-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102810250B (en) | Video based multi-vehicle traffic information detection method | |
| CN105512720B (en) | A kind of public transit vehicle passenger flow statistics method and system | |
| CN103985250B (en) | The holographic road traffic state vision inspection apparatus of lightweight | |
| CN106373394A (en) | Vehicle detection method and system based on video and radar | |
| CN102509089B (en) | Method for recognizing zebra crossing and measuring zebra crossing distance based on line-by-line scanning | |
| CN107066953A (en) | It is a kind of towards the vehicle cab recognition of monitor video, tracking and antidote and device | |
| CN106128121A (en) | Vehicle queue length fast algorithm of detecting based on Local Features Analysis | |
| CN109064495A (en) | A kind of bridge floor vehicle space time information acquisition methods based on Faster R-CNN and video technique | |
| CN106096525A (en) | A kind of compound lane recognition system and method | |
| CN107590999B (en) | Traffic state discrimination method based on checkpoint data | |
| CN104029680A (en) | Lane departure warning system and method based on monocular camera | |
| CN104318781B (en) | Based on the travel speed acquisition methods of RFID technique | |
| CN101571997A (en) | Method and device for fusion processing of multi-source traffic information | |
| CN101615342B (en) | Vehicle detection method | |
| CN103679167A (en) | Method for processing CCD images | |
| CN105118305A (en) | Vehicle management platform of outdoor parking lot exit | |
| CN103310199A (en) | Vehicle model identification method based on high-resolution remote sensing data | |
| CN104504363A (en) | Real-time identification method of sidewalk on the basis of time-space correlation | |
| CN106778540A (en) | Parking detection is accurately based on the parking event detecting method of background double layer | |
| CN103679214A (en) | Vehicle detection method based on online area estimation and multi-feature decision fusion | |
| CN114814823A (en) | Rail vehicle detection system and method based on integration of millimeter wave radar and camera | |
| CN105761504A (en) | Vehicle speed real-time measuring method based on inhomogeneous video image frame collection | |
| CN103295003A (en) | Vehicle detection method based on multi-feature fusion | |
| Yao et al. | Embedded technology and algorithm for video-based vehicle queue length detection | |
| CN105303837A (en) | Method and system for detecting following behavior characteristic parameter of driver |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180209 Termination date: 20180517 |