CN112101268B - Vehicle line pressing detection method based on geometric projection - Google Patents
Vehicle line pressing detection method based on geometric projection Download PDFInfo
- Publication number
- CN112101268B CN112101268B CN202011008149.6A CN202011008149A CN112101268B CN 112101268 B CN112101268 B CN 112101268B CN 202011008149 A CN202011008149 A CN 202011008149A CN 112101268 B CN112101268 B CN 112101268B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- image
- representing
- point
- sub
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 25
- 230000006399 behavior Effects 0.000 claims abstract description 14
- 238000012216 screening Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 13
- 210000005036 nerve Anatomy 0.000 claims description 3
- 238000006386 neutralization reaction Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 abstract description 4
- 238000000605 extraction Methods 0.000 abstract 2
- 238000013527 convolutional neural network Methods 0.000 abstract 1
- 238000003709 image segmentation Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- 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
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- 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
-
- 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
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
Abstract
The invention provides a traffic vehicle line pressing detection method based on geometric projection, which comprises the following steps: image segmentation; carrying out vehicle detection and feature extraction by using a convolutional neural network; screening the detected vehicles; performing chassis fitting on the screened target illegal similar vehicles; and performing line pressing detection judgment on the target illegal similar vehicles. According to the invention, target detection and feature extraction of the vehicle are realized through deep learning, the vehicle chassis fitting is carried out through geometric projection, and then vehicle line pressing judgment is carried out on the basis of the vehicle chassis fitting, so that the vehicle line pressing behavior can be accurately judged.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to a vehicle line pressing detection method based on geometric projection.
Background
In recent years, with the continuous progress of urbanization, the demand of urban traffic is also increased year by year, the quantity of vehicles kept is increased, and the number of traffic illegal events is increased. The traffic violation incident is generally shot by an electronic camera, and a plurality of pictures containing the dynamic process of the illegal action are recorded and synthesized and uploaded to a traffic management platform. In various traffic illegal behaviors, the illegal driving of the vehicle line pressing is easy to cause traffic accidents such as scratch and rear-end collision, so that traffic jam is caused, and serious harm is caused to traffic safety and efficiency. Therefore, more requirements are put forward for the related research of the vehicle line pressing detection.
The most direct method for detecting the vehicle line pressing is to consider the line pressing condition of the vehicle in the illegal picture to be observed, but the labor cost is high, the efficiency is low, and the working time is short. Since a computer-dependent vehicle violation line detection is actually required. The current commonly used vehicle line pressing detection method is that a canny edge detection algorithm and Hough transformation are used for fitting a lane line, and then a vehicle line pressing result is judged according to whether shielding exists on the lane line or not and the existence of line pressing objects, but the algorithm is greatly influenced by the reasons of illumination, weather and shooting angles, the error is large, and the robustness is poor.
The vehicle line pressing detection method related to the invention at present comprises the following steps: the invention patent (publication number: CN110298300A, name: a method for detecting vehicle violation line pressing) discloses a vehicle violation line pressing detection method based on deep learning, but the invention patent does not consider the difference between a vehicle detection frame and a real vehicle chassis frame caused by the visual angle difference, and the invention obtains the vehicle chassis fitting frame by geometric projection, so that the accuracy is higher, the effect is better, and the two have obvious difference.
Disclosure of Invention
The invention provides a traffic vehicle line pressing detection method based on geometric projection, aiming at overcoming the defects of low detection precision, low detection speed and robustness in the prior art.
The technical scheme of the invention is as follows:
a traffic vehicle line pressing detection method based on geometric projection is characterized by comprising the following steps:
step 1: reading the four-in-one vehicle line pressing violation image, dividing the image into four sub-images, and respectively recording the sub-images obtained by dividing the image into four parts, namely the upper left sub-image, the upper right sub-image, the lower left sub-image and the lower right sub-image as I i I1, 2,3,4, image I 4 The middle vehicle is unique and is a target illegal vehicle;
step 2: detecting I by deep convolution nerve detection model i All vehicles in (1), andcomposition set
Wherein the content of the first and second substances,
representing an image I i The envelope box of the jth vehicle in (a),
to represent
The coordinates of the upper left corner of the table,
and
respectively represent
Width and height of (F) i j Representing an image I i The feature vector of the jth vehicle in (1), F i j ＝{f i jk |k＝1,2,...,K}，f i jk Is represented by F i j K-th feature in (1), K representing the vehicle feature vector dimension, m i M represents the number of detected vehicles in the ith sub-image 4 ＝1；
And step 3: screening the set C obtained in the step 2, specifically:
step 3.1: calculating a feature similarity set between the target illegal vehicle and all vehicles by formula (1)
Wherein
Representing an image I 4 Target illegal vehicle and image I i The feature similarity of the jth vehicle in (1),
representing an image I 4 The feature vector of the 1 st vehicle, F i j Representing an image I i K represents a vehicle feature vector dimension,
representing an image I 4 The kth feature of the 1 st vehicle;
step 3.2: respectively obtaining sub-images I according to the feature similarity set S 1 、I 2 、I 3 The vehicle with the highest similarity of the medium features records the envelope frame set as
Representing an image I i The envelope frame of the vehicle with the highest similarity to the target illegal vehicle,
to represent
The coordinates of the upper left corner of the table,
and
respectively represent
Width and height of (d);
and 4, step 4: and (3) performing vehicle chassis fitting operation on the matched vehicle obtained in the step (3), specifically:
step 4.1: calculating the middle points of the vehicle envelope frames in the set B according to the formula (2), wherein the middle points are respectively
Step 4.2: calculating a current vehicle driving direction set G ═ G (G) according to equation (3) i |i＝1,2):
Wherein g is i Representing an image I i Neutralization image I 4 The direction of the vehicle with the highest similarity of the features of the target illegal vehicle,
the center point of the envelope frame of the ith vehicle obtained in the step 4.1 is obtained;
the center point of the envelope frame of the (i + 1) th vehicle obtained in the step 4.1 is obtained;
step 4.3: selecting a vehicle chassis fitting point according to the vehicle direction, which specifically comprises the following steps: if g is i Selecting the upper left corner point and the lower right corner point of the vehicle envelope frame as fitting points when the vehicle envelope frame is larger than or equal to 0; otherwise, if g i If the sum is less than 0, selecting a lower left corner point and an upper right corner point of a vehicle envelope frame as fitting points; respectively marking the selected fitting points as P it (x it ,y it ) And P id (x id ,y id ) (ii) a Wherein P is it (x it ,y it ) Representing the coordinates of the vertices of the i-th sub-image, P, obtained id (x id ,y id ) Representing the coordinates of the lower vertex of the ith sub-image;
step 4.4: according to point P it (x it ,y it ) And a vehicle direction g i Building a straight line l it ：Y＝(g i *θ)X+(y it -g i *θ*x it ) Where θ is a point P it (x it ,y it ) The direction scale factor of (2) is expressed as a straight line l coinciding with the lower boundary of the vehicle envelope d ：Y＝y id Calculating a straight line l it And a straight line l d And the intersection point of (A) is denoted as P id2 (x id2 ,y id2 )；
Step 4.5: according to the calculated point P id (x id ,y id ) Direction g of the vehicle i Building a straight line l id ：Y＝(g i *δ)X+(y id -g i *δ*x id ) Where δ is a point P id (x id ,y id ) The direction scale factor of (2) is expressed as a straight line l coinciding with the upper boundary of the vehicle envelope t ：Y＝y it Calculating a straight line l id And a straight line l t And the intersection point of (A) is denoted as P it2 (x it2 ,y it2 )；
And 4.6, marking the constructed vehicle chassis fitting frame as D ═ P it (x it ,y it ),P id2 (x id2 ,y id2 ),P id (x id ,y id ),P it2 (x it2 ,y it2 )}；
And 5: whether a target illegal vehicle has a line pressing violation behavior is judged according to the vehicle chassis fitting frame D, and the method specifically comprises the following steps: if the vehicle chassis fitting frame D is not intersected with the lane line, the vehicle is represented to have no line pressing illegal behavior; otherwise, calculating the vehicle line-pressing confidence coefficient A according to the formula (4), and when A is larger than lambda, indicating that the vehicle has illegal line-pressing behaviors; when A is less than lambda, the illegal line pressing behavior of the vehicle does not exist;
Wherein S is small Representing the area of the smaller part of the vehicle chassis fitting frame D after being divided by the lane line, S D The area of the vehicle chassis fitting frame D is represented, and the lambda represents a vehicle line pressing confidence threshold value.
Compared with the prior art, the invention has the main beneficial effects that:
the invention provides a vehicle line pressing detection method based on geometric projection, which has strong robustness to environmental change, and realizes higher detection precision due to the fact that the method considers the difference between a vehicle detection frame and a real vehicle chassis frame caused by the difference of visual angles. The method uses deep learning and geometric projection method, so that the human resource cost is greatly reduced, the traffic load and the environmental pollution are reduced, the traffic safety is ensured, and the traffic efficiency is improved.
Drawings
FIG. 1 is a flow chart of the steps of the method of the present invention;
FIG. 2 is a composite image of vehicle line marking violation according to the method of the present invention;
FIG. 3 is a diagram of the effect of the fitting of the vehicle chassis according to the method of the present invention, wherein an opaque quadrangle is a frame of the fitted vehicle chassis.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the detailed description and specific examples, while indicating the scope of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.
As shown in fig. 1, the present embodiment provides a vehicle line pressing detection method based on geometric projection, including the following steps:
step 1: reading the four-in-one vehicle line pressing violation image, dividing the image into four sub-images, and respectively recording the sub-images obtained by dividing the image into four parts, namely the upper left sub-image, the upper right sub-image, the lower left sub-image and the lower right sub-image as I i I1, 2,3,4, image I 4 The middle vehicle is unique and is a target illegal vehicle;
step 2: detecting I by deep convolution nerve detection model i All vehicles in (1), and constitute a set
Wherein the content of the first and second substances,
representing an image I i The envelope box of the jth vehicle in (a),
to represent
The coordinates of the upper left corner of the table,
and
respectively represent
Width and height of (F) i j Representing an image I i The feature vector of the jth vehicle in (1), F i j ＝{f i jk |k＝1,2,...,K}，f i jk Is represented by F i j K-th feature in (1), K representing the vehicle feature vector dimension, m i M represents the number of detected vehicles in the ith sub-image 4 ＝1；
And step 3: screening the set C obtained in the step 2, specifically:
step 3.1: calculating a feature similarity set between the target illegal vehicle and all vehicles by formula (1)
Wherein
Representing an image I 4 Target illegal vehicle and image I i The feature similarity of the jth vehicle in (1),
representing an image I 4 The feature vector of the 1 st vehicle, F i j Representing an image I i K represents a vehicle feature vector dimension,
representing an image I 4 The kth feature of the 1 st vehicle;
step 3.2: respectively obtaining sub-images I according to the feature similarity set S 1 、I 2 、I 3 The vehicle with the highest similarity of the medium features records the envelope frame set as
Representing an image I i The envelope frame of the vehicle with the highest similarity to the target illegal vehicle,
to represent
The coordinates of the upper left corner of the table,
and
respectively represent
Width and height of (d);
and 4, step 4: and (3) performing vehicle chassis fitting operation on the matched vehicle obtained in the step (3), specifically:
step 4.1: calculating the middle points of the vehicle envelope frames in the set B according to the formula (2), wherein the middle points are respectively
Step 4.2: calculating a current vehicle driving direction set G ═ G (G) according to equation (3) i |i＝1,2):
Wherein g is i Representing an image I i Neutralization image I 4 The direction of the vehicle with the highest similarity of the features of the target illegal vehicle,
the center point of the envelope frame of the ith vehicle obtained in the step 4.1 is obtained;
the center point of the envelope frame of the (i + 1) th vehicle obtained in the step 4.1 is obtained;
step 4.3: selecting a vehicle chassis fitting point according to the vehicle direction, which specifically comprises the following steps: if g is i Selecting the upper left corner point and the lower right corner point of the vehicle envelope frame as fitting points when the vehicle envelope frame is larger than or equal to 0; otherwise, if g i If the sum is less than 0, selecting a lower left corner point and an upper right corner point of a vehicle envelope frame as fitting points; respectively marking the selected fitting points as P it (x it ,y it ) And P id (x id ,y id ) (ii) a Wherein P is it (x it ,y it ) Representing the coordinates of the vertices of the i-th sub-image, P, obtained id (x id ,y id ) Representative of the calculated i-thLower vertex coordinates of the subimages;
step 4.4: according to point P it (x it ,y it ) And a vehicle direction g i Building a straight line l it ：Y＝(g i *θ)X+(y it -g i *θ*x it ) Where θ is a point P it (x it ,y it ) The direction scale factor of (2) is expressed as a straight line l coinciding with the lower boundary of the vehicle envelope d ：Y＝y id Calculating a straight line l it And a straight line l d And the intersection point of (A) is denoted as P id2 (x id2 ,y id2 )；
Step 4.5: according to the calculated point P id (x id ,y id ) Direction g of the vehicle i Building a straight line l id ：Y＝(g i *δ)X+(y id -g i *δ*x id ) Where δ is a point P id (x id ,y id ) The direction scale factor of (2) is expressed as a straight line l coinciding with the upper boundary of the vehicle envelope t ：Y＝y it Calculating a straight line l id And a straight line l t And the intersection point of (A) is denoted as P it2 (x it2 ,y it2 )；
And 4.6, marking the constructed vehicle chassis fitting frame as D ═ P it (x it ,y it ),P id2 (x id2 ,y id2 ),P id (x id ,y id ),P it2 (x it2 ,y it2 )}；
And 5: whether a target illegal vehicle has a line pressing violation behavior is judged according to the vehicle chassis fitting frame D, and the method specifically comprises the following steps: if the vehicle chassis fitting frame D is not intersected with the lane line, the vehicle is represented to have no line pressing illegal behavior; otherwise, calculating the vehicle line-pressing confidence coefficient A according to the formula (4), and when A is larger than lambda, indicating that the vehicle has illegal line-pressing behaviors; when A is less than lambda, the illegal line pressing behavior of the vehicle does not exist;
wherein S is small Representing vehicle chassis fitArea of the smaller part of frame D divided by lane line, S D The area of the vehicle chassis fitting frame D is represented, and the lambda represents a vehicle line pressing confidence threshold value.
Claims (1)
1. A traffic vehicle line pressing detection method based on geometric projection is characterized by comprising the following steps:
step 1: reading the four-in-one vehicle line pressing violation image, dividing the image into four sub-images, and respectively recording the sub-images obtained by dividing the image into four parts, namely the upper left sub-image, the upper right sub-image, the lower left sub-image and the lower right sub-image as I i I1, 2,3,4, image I 4 The middle vehicle is unique and is a target illegal vehicle;
step 2: detecting I by deep convolution nerve detection model i All vehicles in (1), and constitute a set
F i j ＞|i＝1,2,3,4；j＝1,2,...,m i And (c) the step of (c) in which,
representing an image I i The envelope box of the jth vehicle in (a),
to represent
The coordinates of the upper left corner of the table,
and
respectively represent
Width and height of (F) i j Representing an image I i J (th) in (2)Feature vector of vehicle, F i j ＝{f i jk |k＝1,2,...,K}，f i jk Is represented by F i j K-th feature in (1), K representing the vehicle feature vector dimension, m i M represents the number of detected vehicles in the ith sub-image 4 ＝1；
And step 3: screening the set C obtained in the step 2, specifically:
step 3.1: calculating a feature similarity set between the target illegal vehicle and all vehicles by formula (1)
Wherein
Representing an image I 4 Target illegal vehicle and image I i The feature similarity of the jth vehicle in (1),
representing an image I 4 The feature vector of the 1 st vehicle, F i j Representing an image I i K represents a vehicle feature vector dimension,
representing an image I 4 The kth feature of the 1 st vehicle;
step 3.2: respectively obtaining sub-images I according to the feature similarity set S 1 、I 2 、I 3 The vehicle with the highest similarity of the medium features records the envelope frame set as
Representing an image I i The envelope frame of the vehicle with the highest similarity to the target illegal vehicle,
to represent
The coordinates of the upper left corner of the table,
and
respectively represent
Width and height of (d);
and 4, step 4: and (3) performing vehicle chassis fitting operation on the matched vehicle obtained in the step (3), specifically:
step 4.1: calculating the middle points of the vehicle envelope frames in the set B according to the formula (2), wherein the middle points are respectively
Step 4.2: calculating a current vehicle driving direction set G ═ G (G) according to equation (3) i |i＝1,2):
Wherein g is i Representing an image I i Neutralization image I 4 The direction of the vehicle with the highest similarity of the features of the target illegal vehicle,
the center point of the envelope frame of the ith vehicle obtained in the step 4.1 is obtained;
the center point of the envelope frame of the (i + 1) th vehicle obtained in the step 4.1 is obtained;
step 4.3: selecting a vehicle chassis fitting point according to the vehicle direction, which specifically comprises the following steps: if g is i The method comprises the following steps that (1) the upper left corner point and the lower right corner point of a vehicle envelope frame are selected to be fit points; otherwise, if g i If the sum is less than 0, selecting a lower left corner point and an upper right corner point of a vehicle envelope frame as fitting points; respectively marking the selected fitting points as P it (x it ,y it ) And P id (x id ,y id ) (ii) a Wherein P is it (x it ,y it ) Representing the coordinates of the vertices of the i-th sub-image, P, obtained id (x id ,y id ) Representing the coordinates of the lower vertex of the ith sub-image;
step 4.4: according to point P it (x it ,y it ) And a vehicle direction g i Constructing equation of straight line l it ：Y＝(g i *θ)X+(y it -g i *θ*x it ) Where θ is a point P it (x it ,y it ) The linear equation of the direction proportional coefficient is recorded as l, and the linear equation is coincident with the lower boundary of the vehicle envelope frame d ：Y＝y id Calculating a straight line l it And a straight line l d And the intersection point of (A) is denoted as P id2 (x id2 ,y id2 )；
Step 4.5: according to the calculated point P id (x id ,y id ) Direction g of the vehicle i Building a straight line l id ：Y＝(g i *δ)X+(y id -g i *δ*x id ) Where δ is a point P id (x id ,y id ) The direction scale factor of (2) is expressed as a straight line l coinciding with the upper boundary of the vehicle envelope t ：Y＝y it Calculating a straight line l id And a straight line l t And the intersection point of (A) is denoted as P it2 (x it2 ,y it2 )；
And 4.6, marking the constructed vehicle chassis fitting frame as D ═ P it (x it ,y it ),P id2 (x id2 ,y id2 ),P id (x id ,y id ),P it2 (x it2 ,y it2 )}；
And 5: whether a target illegal vehicle has a line pressing violation behavior is judged according to the vehicle chassis fitting frame D, and the method specifically comprises the following steps: if the vehicle chassis fitting frame D is not intersected with the lane line, the vehicle is represented to have no line pressing illegal behavior; otherwise, calculating the vehicle line-pressing confidence coefficient A according to the formula (4), and when A is larger than lambda, indicating that the vehicle has illegal line-pressing behaviors; when A is less than lambda, the illegal line pressing behavior of the vehicle does not exist;
Wherein S is small Representing the area of the smaller part of the vehicle chassis fitting frame D after being divided by the lane line, S D The area of the vehicle chassis fitting frame D is represented, and the lambda represents a vehicle line pressing confidence threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011008149.6A CN112101268B (en) | 2020-09-23 | 2020-09-23 | Vehicle line pressing detection method based on geometric projection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011008149.6A CN112101268B (en) | 2020-09-23 | 2020-09-23 | Vehicle line pressing detection method based on geometric projection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112101268A CN112101268A (en) | 2020-12-18 |
| CN112101268B true CN112101268B (en) | 2022-07-29 |
Family
ID=73755905
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011008149.6A Active CN112101268B (en) | 2020-09-23 | 2020-09-23 | Vehicle line pressing detection method based on geometric projection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112101268B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114013456A (en) * | 2021-11-09 | 2022-02-08 | 英博超算(南京)科技有限公司 | Target vehicle screening method for pressing line in automatic driving assistance process |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948418A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of violation guiding based on deep learning |
| CN109948417A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of vehicle based on deep learning drives in the wrong direction illegal automatic auditing method |
| CN109949578A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of vehicle crimping based on deep learning |
| CN110490150A (en) * | 2019-08-22 | 2019-11-22 | 浙江工业大学 | A kind of automatic auditing system of picture violating the regulations and method based on vehicle retrieval |
| CN111402329A (en) * | 2020-03-24 | 2020-07-10 | 上海眼控科技股份有限公司 | Vehicle line pressing detection method and device, computer equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019168869A1 (en) * | 2018-02-27 | 2019-09-06 | Nvidia Corporation | Real-time detection of lanes and boundaries by autonomous vehicles |
-
2020
- 2020-09-23 CN CN202011008149.6A patent/CN112101268B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948418A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of violation guiding based on deep learning |
| CN109948417A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of vehicle based on deep learning drives in the wrong direction illegal automatic auditing method |
| CN109949578A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of vehicle crimping based on deep learning |
| CN110490150A (en) * | 2019-08-22 | 2019-11-22 | 浙江工业大学 | A kind of automatic auditing system of picture violating the regulations and method based on vehicle retrieval |
| CN111402329A (en) * | 2020-03-24 | 2020-07-10 | 上海眼控科技股份有限公司 | Vehicle line pressing detection method and device, computer equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 3D Bounding Box Estimation Using Deep Learning and Geometry;Arsalan Mousavian，et al;《2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)》;20171109;第5632-5640页 * |
| 基于视频图像的车辆违章行为检测识别技术研究;冯荣达;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》;20200815;C034-184 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112101268A (en) | 2020-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109784344B (en) | Image non-target filtering method for ground plane identification recognition | |
| CN104766058B (en) | A kind of method and apparatus for obtaining lane line | |
| US11763485B1 (en) | Deep learning based robot target recognition and motion detection method, storage medium and apparatus | |
| CN111179152B (en) | Road identification recognition method and device, medium and terminal | |
| CN109902676B (en) | Dynamic background-based violation detection algorithm | |
| CN106778659B (en) | License plate recognition method and device | |
| CN103324920A (en) | Method for automatically identifying vehicle type based on vehicle frontal image and template matching | |
| CN109961013A (en) | Recognition methods, device, equipment and the computer readable storage medium of lane line | |
| CN110110608B (en) | Forklift speed monitoring method and system based on vision under panoramic monitoring | |
| CN106485182A (en) | A kind of fuzzy Q R code restored method based on affine transformation | |
| CN111476890B (en) | Method for repairing moving vehicle in three-dimensional scene reconstruction based on image | |
| CN113128507A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN111539303A (en) | Monocular vision-based vehicle driving deviation early warning method | |
| CN112101268B (en) | Vehicle line pressing detection method based on geometric projection | |
| CN114719873A (en) | Low-cost fine map automatic generation method and device and readable medium | |
| CN112801021B (en) | Method and system for detecting lane line based on multi-level semantic information | |
| CN108133231B (en) | Scale-adaptive real-time vehicle detection method | |
| CN106780541B (en) | A kind of improved background subtraction method | |
| CN108846363A (en) | A kind of subregion vehicle bottom shadow detection method based on divergence expression scanning | |
| CN115880683B (en) | Urban waterlogging ponding intelligent water level detection method based on deep learning | |
| CN112950954B (en) | Intelligent parking license plate recognition method based on high-position camera | |
| CN106023270A (en) | Video vehicle detection method based on locally symmetric features | |
| CN116189136A (en) | Deep learning-based traffic signal lamp detection method in rainy and snowy weather | |
| CN111626180B (en) | Lane line detection method and device based on polarization imaging | |
| CN114299247A (en) | Rapid detection and problem troubleshooting method for road traffic sign lines |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A geometric projection based method for vehicle line pressing detection Effective date of registration: 20231127 Granted publication date: 20220729 Pledgee: Zhejiang Lishui Liandu Rural Commercial Bank Co.,Ltd. Pledgor: ZHEJIANG HAOTENG ELECTRON TECHNOLOGY CO.,LTD. Registration number: Y2023980067610 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |