CN103455820A - Method and system for detecting and tracking vehicle based on machine vision technology - Google Patents
Method and system for detecting and tracking vehicle based on machine vision technology Download PDFInfo
- Publication number
- CN103455820A CN103455820A CN201310288011XA CN201310288011A CN103455820A CN 103455820 A CN103455820 A CN 103455820A CN 201310288011X A CN201310288011X A CN 201310288011XA CN 201310288011 A CN201310288011 A CN 201310288011A CN 103455820 A CN103455820 A CN 103455820A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- detection
- road
- tracking
- machine vision
- 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
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method for detecting and tracking a vehicle based on the machine vision technology. The method comprises the step of detecting a structuralized road, wherein a picture is preprocessed, the edges of the picture are enhanced and are mediated in a linear mode, the edges of the road are extracted, and the road region is determined; the step of detecting the moving vehicle, wherein the vehicle in the determined road region is initially recognized, the region of interest is further diminished, and whether a vehicle to be detected exists in the determined road region or not is recognized for the second time; the step of tracking the moving vehicle, wherein the accurate position of the target vehicle is determined through repeated iteration tracking based on the mean value drifting algorithm. The invention further discloses a system for detecting and tracking the vehicle based on the machine vision technology. The vehicle can be distinguished more effectively based on the maximum variance threshold value extraction method, the vehicle can be detected more accurately based on the fusion character algorithm, the region of interest is determined, the secondary shrunk ROI region is detected through an adaboost classifier, the operand is greatly reduced, the real-time performance is improved, and meanwhile the accuracy that the system detects the vehicle is improved.
Description
Technical field
The present invention relates to image processing field, particularly relate to a kind of vehicle detection and tracking and system based on machine vision technique.
Background technology
We study intelligent vehicle, and most crucial problem is exactly how to allow them can the perception surrounding environment, and information extraction, distinguish the place ahead barrier, and utilize these information to complete navigation task.By vision guided navigation Information Monitoring again, therefore than other sensors, more obvious advantage is arranged on quantity of information and picking rate, vision guided navigation not only can gather the position of target in surrounding environment on the one hand, and comprise the external profile of target, color characteristic, textural characteristics, edge feature etc., these are all that other sensors are incomparable; Adopt on the other hand vision guided navigation can well detect the edge of road, and some artificial warning signs on road side, the very important road information of road sign etc., and the scene that presents of camera also meets the rule in the human knowledge world, first from the vision subjective understanding, more further extract other information.
The method of lot of domestic and foreign machine vision realizes Intelligent Recognition at present, adopts and utilizes vehicle movement Characteristics Detection vehicle; Employing utilizes vehicle characteristics to detect vehicle; Employing detects vehicle based on statistical learning; Adopt stereoscopic vision to detect the method for vehicle.Adopt single method, certainly will can not effectively detect vehicle, along with computer hardware software raising, effective integration several different methods how, many vehicle detection trend that is inevitable.
Summary of the invention
Goal of the invention: for problems of the prior art and deficiency, the invention provides a kind of vehicle detection and tracking and system of machine vision technique.
Technical scheme: the vehicle detection based on machine vision technique and tracking comprise the steps:
The detection of structured road: at first picture is carried out to pre-service, then carries out the edge enhancing, and after by straight line, mediate, extract road edge, determine road area;
The detection of moving vehicle: in definite road area, vehicle is identified for the first time, further dwindles area-of-interest, then carries out secondary identification and determines whether to contain the detection vehicle;
The tracking of moving vehicle: by mean shift algorithm, iterate and follow the tracks of the accurate location of determining target vehicle.
Vehicle detection based on machine vision technique and tracker, described system comprises:
The Road Detection unit: for picture is carried out to pre-service, then carry out the edge enhancing, and after by straight line, mediate, extract road edge, determine road area;
Vehicle detection unit: at described definite road area, vehicle being identified for the first time, further dwindle area-of-interest, then carry out secondary identification and determine whether to contain the detection vehicle;
Vehicle tracking unit: for by mean shift algorithm, iterate and follow the tracks of the accurate location of determining target vehicle.
The present invention adopts technique scheme, there is following beneficial effect: compared with prior art, provided by the present inventionly based on machine vision technique, vehicle is carried out to the detection and tracking method, adopt the maximum kind variance to extract the method for threshold value, the more effective differentiation vehicle of energy, make the fusion feature algorithm can detect more accurately vehicle, determine area-of-interest, then in ROI zone secondary dwindled by the adaboost sorter, detect, operand is significantly reduced, improve real-time, improved the accuracy of system to vehicle detection simultaneously.
The accompanying drawing explanation
The structured flowchart that Fig. 1 is the embodiment of the present invention;
Fig. 2 is Road Detection process flow diagram in the embodiment of the present invention;
Fig. 3 is detection flow for the automobile figure in the embodiment of the present invention;
The distribution that Fig. 4 is sample image eigenwert in the embodiment of the present invention;
The final detection that Fig. 5 is vehicle in the embodiment of the present invention.
Embodiment
Below in conjunction with specific embodiment, further illustrate the present invention, should understand these embodiment only is not used in and limits the scope of the invention for the present invention is described, after having read the present invention, those skilled in the art all fall within the application's claims limited range to the modification of the various equivalent form of values of the present invention.
As described in Figure 1, in order to realize, based on machine vision technique, vehicle is carried out to the detection and tracking method, by the input original image is carried out to pre-service, road is detected, determined road area, then in definite ROI zone, the Preliminary detection vehicle, further dwindle the ROI zone, then in this zone, use sorter to be detected vehicle, finally utilize mean shift algorithm to be followed the tracks of target vehicle.
As described in Figure 2, in order to realize the detection to road: 1. at first will carry out pre-service to image, adopt Bright Block algorithm to realize the white balance processing, eliminate the impact of some noises by spatial filtering, after verification experimental verification, adopt medium filtering to carry out smoothing processing, effectively the filtering salt-pepper noise; 2. determine effective coverage, by the camera calibration technology, the zone that in artificial filtering image, some skies etc. need not detect, improve the real-time of operation, and image is carried out to the gray processing processing, reduce the time that image is processed, and prepare for follow-up road edge extracts; 3. through comparative analysis and verification experimental verification, adopt the sobel algorithm that is applicable to native system to be strengthened image border, detect road edge.4. by Hough change detection straight line, determine the road surface scope, judge accurately zone, non-road surface and zone, road surface.When the follow-up detection to vehicle, just can only in road area, detect like this real-time of the operation of raising system greatly.
As described in Figure 3, in order to realize the detection to vehicle: at first in definite road area, priori by vehicle characteristics is carried out secondary-confirmation to vehicle, comprise shadow character, vehicle shape feature, vehicle edge horizontal vertical feature, car plate and taillight feature, further dwindle area-of-interest, left backly utilize the cascade classifier that off-line training is good, complete the final detection to vehicle.Wherein the training of sorter can be divided into off-line training and on-line testing, because off-line training is to carry out under online, utilizes vehicle sample and non-vehicle sample, extracts the Harr feature, training classifier.Detect when system is moved online, carry out in real time, so can not consume for a long time, this cover theoretical system of empirical tests can be good at detecting vehicle.
In order to realize the tracking to vehicle: utilize processing input picture is carried out above, accurately identify the vehicle target in image, the initial position at registration of vehicle center, calculate the probability density distribution of target vehicle color characteristic histogram; Then read the next frame image, and utilize mean shift algorithm, the probability density distribution of the calculated candidate that iterates target area color characteristic histogram; Finally by the Bhattacharyya coefficient, calculate similarity, find out the candidate region model the highest with the vehicle target similarity ，Gai position, position be the next frame image in the position at vehicle target place.
The present embodiment also comprises a kind of vehicle detection based on machine vision technique and tracker that realizes said method, and this system comprises:
The Road Detection unit: for picture is carried out to pre-service, then carry out the edge enhancing, and after by straight line, mediate, extract road edge, determine road area;
Vehicle detection unit: at described definite road area, vehicle being identified for the first time, further dwindle area-of-interest, then carry out secondary identification and determine whether to contain the detection vehicle;
Vehicle tracking unit: for by mean shift algorithm, iterate and follow the tracks of the accurate location of determining target vehicle.
As described in Figure 4, the statistical distribution histogram of vehicle textural characteristics value, added up respectively 128 vehicle class sample entropy and non-vehicle class sample entropy, statistical picture is as Fig. 4, the size that horizontal ordinate is entropy, the ratio that the number that ordinate is sample is shared, wherein minimum entropy is 0.3926, and maximum entropy is 4.814, and average entropy is 2.664, after threshold value 30 deciles, the distribution in whole matrix as shown in the figure.The known sample of interpreting blueprints presents two peak values in 1.5 left and right and 3.7 left and right respectively, is respectively the areal concentration territory of non-vehicle sample and vehicle sample, and this also meets the universal law that our statistics is normal distribution.After sample data is used to the calculating of OTSU algorithm, calculating optimal threshold is 3.04, and this threshold value of empirical tests can be distinguished vehicle and non-vehicle by more effective passing threshold, than the simple judgement by empirical value, has higher accuracy.
Utilize the OTSU algorithm to determine that the optimal threshold concrete grammar of distinguishing vehicle is as follows: in order better to differentiate vehicle, this paper proposes a kind of threshold value of differentiating vehicle of determining based on the OTSU algorithm, and this algorithm is called again Otsu threshold algorithm or maximum between-cluster variance algorithm.Originally be for the analysis image feature, definite threshold is classified, and isolates the algorithm of object from image, is used for distinguishing vehicle and pseudo-vehicle in native system.This paper be take the threshold size of determining entropy and is the example explanation, the numerous vehicle samples that extract, and non-vehicle, and the textural characteristics value of pseudo-vehicle, form the matrix of a n*n, utilizes the method for maximum kind variance to carry out definite threshold, distinguishes more accurately vehicle.Specific algorithm is as follows, has or not vehicle to carry out following formula by following threshold process.
Work as f
t(x, y) value of operation result is to mean that vehicle is arranged at 1 o'clock, and value is to mean in picture not have vehicle, the author to select techniques of discriminant analysis to determine optimal threshold t at 0 o'clock, techniques of discriminant analysis is determined the criterion of optimal threshold, is the inter-class variance maximum between the sample that makes to carry out separating after threshold process.Techniques of discriminant analysis only needs 0 rank square and the 1 rank square of compute histograms.If the sample vehicle fleet size is N, maximal value and the minimum value of statistical sample vehicle texture value, be divided into 250 grades by sample, and the entropy normalizing is assigned to each grade, and the sample number that grade is i is N
i,, to 0 rank square of the sample entropy grade sample distribution that is K, first moment is respectively:
When K=L-1, ω (L-1)=1; μ (L-1)=μ
t, μ
tthe average entropy that is called sample.Be provided with M-1 threshold value: 0≤k
1<k
2<... .k
m-1≤ L-1.The class C that all samples is divided into to M entropy
j(C
j∈ [k
j-1+ 1 ... .k
j]; J=1,2, L, M; k
0=0, k
m=L), all kinds of C
jthe probability ω occurred
jwith mean value be μ
j
ω (0)=0 wherein; μ (0)=0, can obtain thus all kinds of inter-class variances and be
Make the sets of threshold values (k of the value maximum of above formula
1, k
2... ..k
m-1), as the optimal threshold group of M value.Getting M is 2, is divided into 2 classes, can obtain thus the optimal threshold of judgement vehicle.
As shown in subordinate list 1, after determining optimal threshold, the numerical value of statistical sample, the average W that calculates the vehicle sample is E
y, non-vehicle class sample average is E
n, population mean is E, vehicle class class internal variance is V
y, non-vehicle class class internal variance is V
n, total class internal variance is V, two class inter-class variances are V
b, population variance is V
tseparability basis for estimation:
The shared proportion in the fusion feature algorithm that we can obtain textural characteristics thus is 2.31, and the weighted value that in like manner calculates the provincial characteristics of the shape facility of vehicle and shade is 3.51 and 4.13.The weights that each feature is set accordingly are α=0.22, β=0.35, γ=0.41.Last fusion feature computing formula is:
V＝0.22*D+0.35*F+0.41*S
By a large amount of testing authentications, statistical study, we can obtain, when the value of W is 0.55, can distinguishing preferably vehicle and non-vehicle, therefore can carry out Preliminary detection to vehicle by calculating the fusion feature value in native system, and further dwindle area-of-interest.
As described in Figure 5, the effect detected in order to obtain better testing vehicle, we by Feature Fusion Algorithm and and the cascade classifier algorithm combine, at first determine in image whether contain target by Fusion Features, further dwindle area-of-interest, then utilize the cascade classifier trained further to be detected.The image of having selected 500 frames to contain vehicle, Fig. 5 is typical four width images wherein.Wherein (a) is (b) detection under intense light irradiation, (c) (d) be the low light level according under detection, 200 meters testing results with interior vehicle target of actual range are as shown, when the value of V is greater than 0.55, we just judge that it is vehicle zone to be checked, further applies the cascade classifier calculated.The result detected is as subordinate list 2, and experimental data shows, when two kinds of algorithms in conjunction with after, when light changes, native system still has very high verification and measurement ratio, proves that this algorithm has very high Stability and veracity.
| To be measured | E y | E n | E | V y | V n | V | V B | V T |
| Numerical value | 3.4881 | 1.6903 | 2.5793 | 0.001733 | 0.00165 | 0.00169 | 0.00202 | 0.00371 |
| Environment | Actual vehicle | The correct number that detects | The flase drop vehicle | Verification and measurement ratio | False drop rate |
| Light a little less than | 235 | 221 | 3 | 94.0% | 1.3% |
| Light is stronger | 287 | 274 | 5 | 95.5% | 1.7% |
| Normal road conditions | 312 | 302 | 4 | 96.8% | 1.2% |
Claims (8)
1. the vehicle detection based on machine vision technique and tracking, is characterized in that, comprises the steps:
The detection of structured road: at first picture is carried out to pre-service, then carries out the edge enhancing, and after by straight line, mediate, extract road edge, determine road area;
The detection of moving vehicle: in definite road area, vehicle is identified for the first time, further dwindles area-of-interest, then carries out secondary identification and determines whether to contain the detection vehicle;
The tracking of moving vehicle: by mean shift algorithm, iterate and follow the tracks of the accurate location of determining target vehicle.
2. vehicle detection and the tracking based on machine vision technique as claimed in claim 1, it is characterized in that: in the detecting step of described structured road, adopt white balance and medium filtering to carry out the image pre-service, adopt the sobel algorithm to carry out the edge enhancing to image, by Hough transfer pair image, carry out the straight line kneading.
3. vehicle detection and the tracking based on machine vision technique as claimed in claim 1, it is characterized in that: in the detecting step of described moving vehicle, adopt the method for Fusion Features to identify for the first time vehicle, further dwindle area-of-interest, when the horizontal vertical features in textural characteristics, edge, the shadow character of vehicle are judged, quote the OTSU algorithm, determine the threshold value that image feature value is judged, good Adaboost sorter completes the detection to vehicle finally to utilize off-line training.
4. vehicle detection and the tracking based on machine vision technique as claimed in claim 1, it is characterized in that: in the tracking step of described moving vehicle, utilize the statistic histogram of color characteristic to describe the target vehicle model, similarity degree by Bhattacharyya coefficient calculations auto model between consecutive frame, by the accurate location of definite target vehicle that iterates.
5. the vehicle detection based on machine vision technique and tracker, is characterized in that, described system comprises:
The Road Detection unit: for picture is carried out to pre-service, then carry out the edge enhancing, and after by straight line, mediate, extract road edge, determine road area;
Vehicle detection unit: at described definite road area, vehicle being identified for the first time, further dwindle area-of-interest, then carry out secondary identification and determine whether to contain the detection vehicle;
Vehicle tracking unit: for by mean shift algorithm, iterate and follow the tracks of the accurate location of determining target vehicle.
6. vehicle detection and the tracker based on machine vision technique as claimed in claim 5, it is characterized in that: in described Road Detection unit, adopt white balance and medium filtering to carry out the image pre-service, adopt the sobel algorithm to carry out the edge enhancing to image, by Hough transfer pair image, carry out the straight line kneading.
7. vehicle detection and the tracker based on machine vision technique as claimed in claim 5, it is characterized in that: in described vehicle detection unit, adopt the method for Fusion Features to identify for the first time vehicle, further dwindle area-of-interest, when the horizontal vertical features in textural characteristics, edge, the shadow character of vehicle are judged, quote the OTSU algorithm, determine the threshold value that image feature value is judged, good Adaboost sorter completes the detection to vehicle finally to utilize off-line training.
8. vehicle detection and the tracker based on machine vision technique as claimed in claim 5, it is characterized in that: in described vehicle tracking unit, utilize the statistic histogram of color characteristic to describe the target vehicle model, similarity degree by Bhattacharyya coefficient calculations auto model between consecutive frame, by the accurate location of definite target vehicle that iterates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310288011XA CN103455820A (en) | 2013-07-09 | 2013-07-09 | Method and system for detecting and tracking vehicle based on machine vision technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310288011XA CN103455820A (en) | 2013-07-09 | 2013-07-09 | Method and system for detecting and tracking vehicle based on machine vision technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103455820A true CN103455820A (en) | 2013-12-18 |
Family
ID=49738162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310288011XA Pending CN103455820A (en) | 2013-07-09 | 2013-07-09 | Method and system for detecting and tracking vehicle based on machine vision technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103455820A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463104A (en) * | 2014-11-14 | 2015-03-25 | 武汉工程大学 | Fast detecting method and device for static vehicle target |
| CN104732235A (en) * | 2015-03-19 | 2015-06-24 | 杭州电子科技大学 | Vehicle detection method for eliminating night road reflective interference |
| CN104899898A (en) * | 2015-05-28 | 2015-09-09 | 华南理工大学 | Multidimensional information probability model based road surface detection method |
| CN105138987A (en) * | 2015-08-26 | 2015-12-09 | 电子科技大学 | Vehicle detection method based on aggregation channel characteristic and motion estimation |
| CN106228106A (en) * | 2016-06-27 | 2016-12-14 | 开易（北京）科技有限公司 | The real-time vehicle detection filter method of a kind of improvement and system |
| CN106372619A (en) * | 2016-09-20 | 2017-02-01 | 北京工业大学 | Vehicle robustness detection and divided-lane arrival accumulative curve estimation method |
| CN106446790A (en) * | 2016-08-30 | 2017-02-22 | 上海交通大学 | Method for tracking and analyzing traffic video flow of fixed camera |
| CN106845493A (en) * | 2016-12-06 | 2017-06-13 | 西南交通大学 | The identification at railroad track close-range image rail edge and matching process |
| CN107274668A (en) * | 2016-04-06 | 2017-10-20 | 北京瑞拓电子技术发展有限公司 | A kind of congestion in road modeling method based on vehicle detection |
| CN109034267A (en) * | 2018-08-20 | 2018-12-18 | 张亮 | Piece caudal flexure intelligent selecting method |
| CN109190523A (en) * | 2018-08-17 | 2019-01-11 | 武汉大学 | A kind of automobile detecting following method for early warning of view-based access control model |
| CN109740595A (en) * | 2018-12-27 | 2019-05-10 | 武汉理工大学 | A kind of oblique moving vehicles detection and tracking system and method based on machine vision |
| CN110677125A (en) * | 2019-10-11 | 2020-01-10 | 国网冀北电力有限公司秦皇岛供电公司 | Arc fault detection method and device |
| CN111797701A (en) * | 2020-06-10 | 2020-10-20 | 东莞正扬电子机械有限公司 | Road obstacle sensing method and system for vehicle multi-sensor fusion system |
| CN112365448A (en) * | 2020-10-20 | 2021-02-12 | 天津大学 | Fabric defect detection method in warp knitting process |
| CN113822447A (en) * | 2021-09-23 | 2021-12-21 | 深圳市星卡科技有限公司 | Intelligent detection method and device for automobile maintenance and vehicle-mounted equipment |
| CN114610029A (en) * | 2022-03-09 | 2022-06-10 | 奇瑞万达贵州客车股份有限公司 | Intelligent vehicle vision system with autonomous navigation function |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1897015A (en) * | 2006-05-18 | 2007-01-17 | 王海燕 | Method and system for inspecting and tracting vehicle based on machine vision |
| US20070263901A1 (en) * | 2006-05-12 | 2007-11-15 | National Chiao Tung University | Real-time nighttime vehicle detection and recognition system based on computer vision |
| CN101193314A (en) * | 2006-11-30 | 2008-06-04 | 北京思比科微电子技术有限公司 | Image processing device and method for image sensor |
-
2013
- 2013-07-09 CN CN201310288011XA patent/CN103455820A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070263901A1 (en) * | 2006-05-12 | 2007-11-15 | National Chiao Tung University | Real-time nighttime vehicle detection and recognition system based on computer vision |
| CN1897015A (en) * | 2006-05-18 | 2007-01-17 | 王海燕 | Method and system for inspecting and tracting vehicle based on machine vision |
| CN101193314A (en) * | 2006-11-30 | 2008-06-04 | 北京思比科微电子技术有限公司 | Image processing device and method for image sensor |
Non-Patent Citations (3)
| Title |
|---|
| 胡会珍: "基于视觉的运动车辆识别与跟踪方法研究", 《信息科技辑》 * |
| 陈烁华: "Sobel-Hough 检测方法在机器视觉倒车辅助系统的应用研究", 《机械设计与制造》 * |
| 龚树锋: "基于机器视觉的道路检测与车辆检测算法研究", 《信息科技辑》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463104A (en) * | 2014-11-14 | 2015-03-25 | 武汉工程大学 | Fast detecting method and device for static vehicle target |
| CN104463104B (en) * | 2014-11-14 | 2018-06-08 | 武汉工程大学 | A kind of stationary vehicle target rapid detection method and device |
| CN104732235A (en) * | 2015-03-19 | 2015-06-24 | 杭州电子科技大学 | Vehicle detection method for eliminating night road reflective interference |
| CN104732235B (en) * | 2015-03-19 | 2017-10-31 | 杭州电子科技大学 | A kind of vehicle checking method for eliminating the reflective interference of road at night time |
| CN104899898A (en) * | 2015-05-28 | 2015-09-09 | 华南理工大学 | Multidimensional information probability model based road surface detection method |
| CN104899898B (en) * | 2015-05-28 | 2018-01-05 | 华南理工大学 | Pavement detection method based on multidimensional information probabilistic model |
| CN105138987B (en) * | 2015-08-26 | 2018-05-18 | 电子科技大学 | A kind of vehicle checking method based on converging channels feature and estimation |
| CN105138987A (en) * | 2015-08-26 | 2015-12-09 | 电子科技大学 | Vehicle detection method based on aggregation channel characteristic and motion estimation |
| CN107274668A (en) * | 2016-04-06 | 2017-10-20 | 北京瑞拓电子技术发展有限公司 | A kind of congestion in road modeling method based on vehicle detection |
| CN106228106A (en) * | 2016-06-27 | 2016-12-14 | 开易（北京）科技有限公司 | The real-time vehicle detection filter method of a kind of improvement and system |
| CN106228106B (en) * | 2016-06-27 | 2019-05-10 | 开易（北京）科技有限公司 | A kind of improved real-time vehicle detection filter method and system |
| CN106446790A (en) * | 2016-08-30 | 2017-02-22 | 上海交通大学 | Method for tracking and analyzing traffic video flow of fixed camera |
| CN106372619A (en) * | 2016-09-20 | 2017-02-01 | 北京工业大学 | Vehicle robustness detection and divided-lane arrival accumulative curve estimation method |
| CN106372619B (en) * | 2016-09-20 | 2019-08-09 | 北京工业大学 | A kind of detection of vehicle robust and divided lane reach summation curve estimation method |
| CN106845493A (en) * | 2016-12-06 | 2017-06-13 | 西南交通大学 | The identification at railroad track close-range image rail edge and matching process |
| CN109190523B (en) * | 2018-08-17 | 2021-06-04 | 武汉大学 | Vehicle detection tracking early warning method based on vision |
| CN109190523A (en) * | 2018-08-17 | 2019-01-11 | 武汉大学 | A kind of automobile detecting following method for early warning of view-based access control model |
| CN109034267A (en) * | 2018-08-20 | 2018-12-18 | 张亮 | Piece caudal flexure intelligent selecting method |
| CN109740595A (en) * | 2018-12-27 | 2019-05-10 | 武汉理工大学 | A kind of oblique moving vehicles detection and tracking system and method based on machine vision |
| CN109740595B (en) * | 2018-12-27 | 2022-12-30 | 武汉理工大学 | Oblique vehicle detection and tracking system and method based on machine vision |
| CN110677125A (en) * | 2019-10-11 | 2020-01-10 | 国网冀北电力有限公司秦皇岛供电公司 | Arc fault detection method and device |
| CN110677125B (en) * | 2019-10-11 | 2022-07-01 | 国网冀北电力有限公司秦皇岛供电公司 | Arc fault detection method and device |
| CN111797701A (en) * | 2020-06-10 | 2020-10-20 | 东莞正扬电子机械有限公司 | Road obstacle sensing method and system for vehicle multi-sensor fusion system |
| CN111797701B (en) * | 2020-06-10 | 2024-05-24 | 广东正扬传感科技股份有限公司 | Road obstacle sensing method and system for vehicle multi-sensor fusion system |
| CN112365448A (en) * | 2020-10-20 | 2021-02-12 | 天津大学 | Fabric defect detection method in warp knitting process |
| CN113822447A (en) * | 2021-09-23 | 2021-12-21 | 深圳市星卡科技有限公司 | Intelligent detection method and device for automobile maintenance and vehicle-mounted equipment |
| CN114610029A (en) * | 2022-03-09 | 2022-06-10 | 奇瑞万达贵州客车股份有限公司 | Intelligent vehicle vision system with autonomous navigation function |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103455820A (en) | Method and system for detecting and tracking vehicle based on machine vision technology | |
| CN102289660B (en) | Method for detecting illegal driving behavior based on hand gesture tracking | |
| CN108830188B (en) | Vehicle detection method based on deep learning | |
| KR100912746B1 (en) | Method for traffic sign detection | |
| CN101859382B (en) | License plate detection and identification method based on maximum stable extremal region | |
| CN102567744B (en) | Method for determining quality of iris image based on machine learning | |
| CN103198303B (en) | A kind of gender identification method based on facial image | |
| Zhang et al. | Study on traffic sign recognition by optimized Lenet-5 algorithm | |
| CN102254428B (en) | Traffic jam detection method based on video processing | |
| CN105205486A (en) | Vehicle logo recognition method and device | |
| CN107194393B (en) | Method and device for detecting temporary license plate | |
| CN103927548B (en) | Novel vehicle collision avoiding brake behavior detection method | |
| CN103048329A (en) | Pavement crack detecting method based on active contour model | |
| CN104182985A (en) | Remote sensing image change detection method | |
| CN104183142A (en) | Traffic flow statistics method based on image visual processing technology | |
| CN104077594A (en) | Image recognition method and device | |
| CN103839033A (en) | Face identification method based on fuzzy rule | |
| CN103971106A (en) | Multi-view human facial image gender identification method and device | |
| CN105046218A (en) | Multi-feature traffic video smoke detection method based on serial parallel processing | |
| CN102169631A (en) | Manifold-learning-based traffic jam event cooperative detecting method | |
| CN105225523A (en) | A kind of parking space state detection method and device | |
| Kuang et al. | Real-Time Detection and Recognition of Road Traffic Signs using MSER and Random Forests. | |
| CN111553214A (en) | Method and system for detecting smoking behavior of driver | |
| CN104915642A (en) | Method and apparatus for measurement of distance to vehicle ahead | |
| CN106557740A (en) | The recognition methods of oil depot target in a kind of remote sensing images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131218 |