CN105809099A - Safety belt detection method based on monitor image - Google Patents
Safety belt detection method based on monitor image Download PDFInfo
- Publication number
- CN105809099A CN105809099A CN201410855403.4A CN201410855403A CN105809099A CN 105809099 A CN105809099 A CN 105809099A CN 201410855403 A CN201410855403 A CN 201410855403A CN 105809099 A CN105809099 A CN 105809099A
- Authority
- CN
- China
- Prior art keywords
- straight line
- image
- pixel
- detection method
- method based
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a safety belt detection method based on a monitor image. The method comprises the steps that the monitor image to be detected is acquired; a license plate in the image is identified, and the position area of the driver of a vehicle is determined according to the position of the license plate; the image of the position area of the driver is intercepted; and whether the intercepted image contains a safety belt image is detected. According to the invention, image identifying and analyzing are carried out on the area of the driver of the vehicle in the monitor video image; the scope of image identifying and analyzing is narrowed; the detection efficiency of the wearing state of a safety belt can be effectively improved; after grey processing is carried out on the video monitor image, a line corresponding to a safety belt edge image is detected; the image quality requirement of the monitor video image is reduced; monitor images of different pixel sizes and definition can be accurately detected; and the general applicability of the scheme is enhanced.
Description
Technical field
The present invention relates to image recognition and mode identification technology, particularly relate to a kind of Safe belt detection method based on monitoring image.
Background technology
Along with social economy quickly increases, the living standard of people is improved constantly, auto industry high speed development simultaneously, increasing family selection automobile is as the vehicles ridden instead of walk and go on a journey, modern trip mode, while provide comfortable choice for traveling easily for people, there is also owing to driver or rider violate accident and the potential safety hazard that road traffic laws and regulations brings.
Whether vehicle travel process is worn seat belt and is directly connected to the life security of driver and crew, when collision occurs, in the seat belt spatial dimension that can be fixed on around seat by driver and crew certain, thus avoiding occurring to clash into or be spilled over the outer grievous injury caused of car with object in car because health is out of control.The law on road traffic safety rule clear stipulaties that China is existing, during motor-driven vehicle going, driver and passenger should use according to regulations seat belt.
At present, road traffic monitoring system has become increasingly popular and perfect, exactly the vehicle travelled on road can be shot, obtaining video image clearly, the infringement not wearing seat belt by regulation for driver and crew in vehicle being travelled by road traffic monitoring system confirms to provide necessary basic condition.What currently mainly adopt is road traffic law violation behavior is judged by the video image of shooting is carried out the mode of manual detection, and owing to video monitoring image data volume is huge, manually-operated inefficiency, accuracy is not high.Therefore, how video monitoring image is carried out automatic, efficiently, identifying processing intelligently, detect the unlawful practice not wearing seat belt rapidly and accurately, become technical problem urgently to be resolved hurrily.
Summary of the invention
In order to solve above-mentioned technical problem at least one, it is an object of the invention to provide a kind of Safe belt detection method based on monitoring image, analyze by monitoring image is identified, it is possible to accurately whether vehicle operator is worn seat belt and carries out detection and judge.
In view of this, the present invention provides a kind of Safe belt detection method based on monitoring image, including: obtain monitoring image to be detected；Identify the car plate in described image, determine the vehicle drive people band of position according to described car plate position；Intercept the image of the described driver band of position；Detect and whether described truncated picture comprises seat belt image.
Preferably, car plate in the described image of described identification, the step of the vehicle drive people band of position is determined according to described car plate position, it is specially, identify the car plate in described image, with the position, the upper right corner in license plate image region for end points, the region choosing predetermined width and preset height is the vehicle drive people band of position.
Preferably, before whether the described truncated picture of described detection comprises the step of seat belt image, also include, the driver band of position image of described intercepting is carried out Face datection, according to the face location detected, intercept the image of driver's upper body torso area.
Preferably, whether the described truncated picture of described detection comprises the step of seat belt image, specifically, described truncated picture to be carried out image enhaucament and edge extracting, obtain and described image meets pre-conditioned straight line set；Straight line in described straight line set is detected, it is determined whether there is the straight line corresponding with described seat belt image border.
Preferably, described described truncated picture is carried out image enhaucament and edge extracting, obtain the step meeting pre-conditioned straight line set in described image, it is specially, described truncated picture carries out gray proces, image filtering, rim detection and/or noise filtering process, detect and obtain the straight line set meeting default slope in described truncated picture.
Preferably, the straight line in described straight line set comprises at least 10 pixels, and the slope of described straight line is more than 0.5 and less than 3.7.
Preferably, described to set in straight line detect, determine whether there is the step of the straight line corresponding with described seat belt image border, specifically, each pixel calculated respectively on described straight line set cathetus in the horizontal direction or vertical direction each side choose the variance yields T of the quadrilateral area that m pixel is formed at described straight line；Calculate each pixel on described straight line in the horizontal direction or vertical direction on the left of described straight line, choose the variance yields L of the quadrilateral area that n pixel is formed, and calculate each pixel on described straight line in the horizontal direction or vertical direction on the right side of described straight line, choose the variance yields R of the quadrilateral area that n pixel is formed, wherein n is more than m；Judge respectively L whether less than T and R whether less than T, if L less than T and/or R less than T, then calculate each pixel on described straight line in the horizontal direction or vertical direction each side choose the variance yields V of the quadrilateral area that n pixel is formed respectively at described straight line；Judge that whether V is more than T, if, it is determined that described straight line is the straight line corresponding with described seat belt image border.
Preferably, the described each pixel calculated respectively on described straight line set cathetus in the horizontal direction or vertical direction before described straight line each side chooses the step of variance yields T of the quadrilateral area that m pixel is formed, also include, choose all of parallel lines in described straight line set, by in described straight line set between two distance be divided into cluster less than the straight line of predetermined threshold value, by distance weighted be straight line by the fitting a straight line in every cluster.
Preferably, the described each pixel calculated respectively on described straight line set cathetus in the horizontal direction or vertical direction before described straight line each side chooses the step of variance yields T of the quadrilateral area that m pixel is formed, also include, the straight line in described straight line set is carried out drift and corrects.
Preferably, described judge that whether V is more than T, if, then determine that step that described straight line is the straight line corresponding with described seat belt image border is particularly as follows: judge that whether V is more than T, if, then by described straight line in the horizontal direction or vertical direction n pixel of translation, each pixel calculated after translating on described straight line in the horizontal direction or vertical direction each side choose the variance yields W of the quadrilateral area that n pixel is formed at described straight line；Judge that whether W is more than T, if, it is determined that described straight line is the straight line that described seat belt image border is corresponding.
The invention provides a kind of Safe belt detection method based on monitoring image, the image-region of vehicle drive people is determined according to the car plate position in traffic monitoring video image, obtain after this area image is processed image meets pre-conditioned straight line set, by traveling through this straight line set, determine whether there is the straight line corresponding with seat belt edge image, if existing, it is determined that in described image, vehicle drive people has worn seat belt.The present invention is by carrying out image recognition analysis to vehicle drive people region in monitor video image, reduce the scope of image recognition analysis, the detection efficiency of wearing safety belt state can be effectively improved, simultaneously, adopt and detect the straight line corresponding with seat belt edge image after video monitoring image is carried out process gray proces, reduce the image quality requirements to monitor video image, the monitoring image of multiple different pixels size Yu definition accurately can be detected, improve the general applicability of scheme.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, the accompanying drawing used required in embodiment or description of the prior art will be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings:
Fig. 1 illustrates the embodiment of the present invention flow chart based on the Safe belt detection method of monitoring image.
Detailed description of the invention
In order to be more clearly understood that the purpose of the present invention, feature and advantage, below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.It should be noted that when not conflicting, embodiments herein and the feature in embodiment can be combined with each other.
Elaborate a lot of detail in the following description so that fully understanding the present invention; but; this is only some embodiments of the present invention; the present invention can also adopt other to be different from other modes described here to implement; therefore, protection scope of the present invention is not by the restriction of following public specific embodiment.
Fig. 1 illustrates the embodiment of the present invention flow chart based on the Safe belt detection method of monitoring image.
The embodiment of the present invention, based on the Safe belt detection method of monitoring image, including step 101, obtains monitoring image to be detected；Step 102, identifies the car plate in image, determines the vehicle drive people band of position according to car plate position；Step 103, intercepts the image of the driver band of position；Step 104, carries out image enhaucament and edge treated to truncated picture, obtains in image and meets pre-conditioned straight line set；Step 105, detects the straight line in straight line set, it is determined whether there is the straight line corresponding with seat belt image border.
In this technical scheme, obtain monitor video image to be detected, wherein comprise the anterior direct picture of vehicle, by identifying that car plate position determined by the car plate in image, according to car plate position and dimension scale, determine the vehicle drive people band of position, specifically, it is possible to position, the license plate image upper right corner for end points, choosing length is 5 car plate width, it is highly that the region of 15 car plate width is as the vehicle drive people band of position, the image of the intercepting driver band of position.
In technique scheme, preferably, in step 104, the driver band of position image intercepted is carried out Face datection, according to the face location detected, analyze and positioning security band detection region, intercept seat belt detection zone area image further, truncated picture is carried out image enhaucament and edge treated, obtains in image and meet pre-conditioned straight line set L1.
In this technical scheme, the driver band of position image intercepted is carried out Face datection, according to the face location region detected, analyze the image-region determining vehicle drive upper part of human body trunk, intercept this area image, truncated picture is carried out image enhaucament and edge treated, obtain in image and meet pre-conditioned straight line set, thus the scope that downscaled images processes further, improve image processing efficiency.
In technique scheme, it is preferable that in step 104, truncated picture is carried out gray proces, image filtering, rim detection and/or noise filtering and processes, obtain in image and meet pre-conditioned straight line set.Specifically: adopt Canny operator that truncated picture is carried out Image Edge-Detection, detect and extract the straight line set L1 meeting default slope in this truncated picture by LSD (LineSegmentDetector) line detection algorithm, straight line in this straight line set comprises at least 10 pixels, and the slope of described straight line is more than 0.5 and less than 3.7.
In this technical scheme, different according to the complexity of video monitoring image, do not wait and close a plurality of straight line apart by the straight line set that step 104 obtains can exist length, described step 104 also includes the screening further to straight line set and optimizes.Specifically: will straight line set L1 be parallel to each other and distance is divided into one bunch less than the straight line of 5 pixels to each other, according to bunch in all straight lines of comprising obtain a new straight line with bunch centre distance by matching after averaging method or method of least square weighting, this straight line is replaced original all straight lines in this bunch, the straight line set L2 after thus being optimized.
In technique scheme, by the image that LSD line detection algorithm detects and extracts meets the straight line of default slope, can there is a degree of pixel-shift, namely the Origin And Destination of straight line is not be completely in borderline region, but in several pixels of certain limit bias internal of borderline region.Described step 104 also includes that the straight line in set is carried out drift and corrects.Specifically: between straight line Origin And Destination, edge is perpendicular to rectilinear direction and chooses each K the pixel in straight line left and right respectively, the arrangement setting described K pixel meets Gauss distribution, calculate local maximum as new straight line starting point and partial points, the all straight lines in L2 set are made all accurately to be distributed in local boundary region, thus the straight line set L3 after being corrected.
In technique scheme, described step 105, first respectively calculate straight line set L3 cathetus A1 Origin And Destination between each pixel in the horizontal direction or vertical direction choose the variance yields T of the quadrilateral area that m pixel is formed respectively in the described straight line left and right sides, wherein m value is 3.Then calculate each pixel on straight line A1 in the horizontal direction or vertical direction on the left of straight line A1, choose the variance yields L of the quadrilateral area that n pixel is formed, and calculate each pixel on straight line A1 in the horizontal direction or vertical direction on the right side of straight line A1, choose the variance yields R of the quadrilateral area that n pixel is formed, if L is less than T, the left field then determining straight line A1 is seat belt image Probability Area, and straight line A1 is arranged a mark Mark_1；If R is less than T, it is determined that the right side area of straight line A1 is seat belt image Probability Area, and straight line A1 is arranged a mark Mark_2；If L is less than T, and R is less than T, it is determined that the region, the left and right sides of straight line A1 is seat belt image Probability Area, and straight line A1 is arranged a mark Mark_3；If L and R is all higher than T, it is determined that the straight line A1 left and right sides is not all seat belt image Probability Area, and this straight line is arranged a mark Mark_0.
In this technical scheme, in traversal straight line set L3 after all of straight line, choose and straight line set L3 identifies not for all straight lines of Mark_0, calculate centered by described straight line on described straight line each pixel in the horizontal direction or vertical direction respectively at the variance yields V each side choosing the quadrilateral area that n pixel is formed of described straight line, if V is more than T, then determine that this straight line is straight line corresponding to seat belt image border, namely the video image of described intercepting exists seat belt image.
In technique scheme, in order to further ensure that the accuracy that seat belt image detects, when judging V more than T, then by described straight line in the horizontal direction or vertical direction translation n pixel, specifically, if described straight line be designated Mark_1, then by described straight line along the horizontal or vertical direction to n pixel of translation on the left of described straight line；If described straight line be designated Mark_1, then by described straight line along the horizontal or vertical direction to n pixel of translation on the right side of described straight line；If described straight line be designated Mark_3, then by described straight line along the horizontal or vertical direction respectively to n pixel of on the left of described straight line and right side translation.After calculating translation each pixel on described straight line in the horizontal direction or vertical direction each side choose the variance yields W of the quadrilateral area that n pixel is formed respectively at described straight line, judge that whether W is more than T, if, it is determined that described straight line is the straight line that described safety image edge is corresponding.
In this technical scheme, by by straight line corresponding for calculated seat belt image border, to seat belt image region translation, and centered by the straight line after translation, calculate the variance yields of straight line two side areas, by multilevel iudge, the straight line corresponding to seat belt image border is further determined that, thus ensureing accuracy and the stability of testing result.
The present invention is based on the Safe belt detection method of monitoring image, by to after in traffic surveillance videos image, the image in vehicle drive people region intercepts, the straight line set meeting preset length and slope condition in image is extracted by gradation of image process, noise filtering etc., straight line in straight line set is detected and determines whether there is the straight line corresponding with seat belt image border, if existing, it is determined that in described image, vehicle drive people has worn seat belt.The present invention is by identifying monitoring image characteristic area, intercept vehicle drive people's image-region to detect, the wearing situation of vehicle operator's seat belt can be detected rapidly and accurately, effectively improve the efficiency for the detection of urban road traffic safety monitoring image and stability.
Again state, all features disclosed in this specification, or the step in disclosed all methods or process, except mutually exclusive feature and/or step, all can combine by any way.
This specification (include any accessory claim, summary and accompanying drawing) disclosed in any feature, unless specifically stated otherwise, all can by other equivalences or there is the alternative features of similar purpose replaced.That is, unless specifically stated otherwise, each feature is an example in a series of equivalence or similar characteristics.
The invention is not limited in aforesaid detailed description of the invention.The present invention can expand to any new feature disclosed in this manual or any new combination, and the step of the arbitrary new method disclosed or process or any new combination.
Claims (10)
1. the Safe belt detection method based on monitoring image, it is characterised in that including:
Obtain monitoring image to be detected；
Identify the car plate in described image, determine the vehicle drive people band of position according to described car plate position；
Intercept the image of the described driver band of position；
Detect and whether described truncated picture comprises seat belt image.
2. the Safe belt detection method based on monitoring image according to claim 1, it is characterised in that the car plate in the described described image of identification, determines the step of the vehicle drive people band of position according to described car plate position, particularly as follows:
Identifying the car plate in described image, with the position, the upper right corner in license plate image region for end points, the region choosing predetermined width and preset height is the vehicle drive people band of position.
3. the Safe belt detection method based on monitoring image according to claim 2, it is characterised in that before whether comprising the step of seat belt image in the described described truncated picture of detection, also include:
The driver band of position image of described intercepting is carried out Face datection, according to the face location detected, intercepts the image of driver's upper body torso area.
4. the Safe belt detection method based on monitoring image according to any one of claim 1 to 3, it is characterised in that whether comprise the step of seat belt image in the described described truncated picture of detection, particularly as follows:
Described truncated picture is carried out image enhaucament and edge extracting, obtains and described image meets pre-conditioned straight line set；
Straight line in described straight line set is detected, it is determined whether there is the straight line corresponding with described seat belt image border.
5. the Safe belt detection method based on monitoring image according to claim 4, it is characterised in that described described truncated picture carries out image enhaucament and edge extracting, obtains the step meeting pre-conditioned straight line set in described image, particularly as follows:
Described truncated picture carries out gray proces, image filtering, rim detection and/or noise filtering process, detect and obtain the straight line set meeting default slope in described truncated picture.
6. the Safe belt detection method based on monitoring image according to claim 5, it is characterised in that the straight line in described straight line set comprises at least 10 pixels, the slope of described straight line is more than 0.5 and less than 3.7.
7. the Safe belt detection method based on monitoring image according to claim 6, it is characterised in that described straight line in described set is detected, it is determined whether there is the step of the straight line corresponding with described seat belt image border, particularly as follows:
Calculate respectively each pixel on described straight line set cathetus in the horizontal direction or vertical direction each side choose the variance yields T of the quadrilateral area that m pixel is formed at described straight line；
Calculate each pixel on described straight line in the horizontal direction or vertical direction on the left of described straight line, choose the variance yields L of the quadrilateral area that n pixel is formed, and calculate each pixel on described straight line in the horizontal direction or vertical direction on the right side of described straight line, choose the variance yields R of the quadrilateral area that n pixel is formed, wherein n is more than m；
Judge respectively L whether less than T and R whether less than T, if L less than T and/or R less than T, then calculate each pixel on described straight line in the horizontal direction or vertical direction each side choose the variance yields V of the quadrilateral area that n pixel is formed respectively at described straight line；
Judge that whether V is more than T, if, it is determined that described straight line is the straight line corresponding with described seat belt image border.
8. the Safe belt detection method based on monitoring image according to claim 7, it is characterized in that, the described each pixel calculated respectively on described straight line set cathetus in the horizontal direction or vertical direction before described straight line each side chooses the step of variance yields T of the quadrilateral area that m pixel is formed, also include:
Choose all of parallel lines in described straight line set, by described straight line set between two distance be divided into cluster less than the straight line of predetermined threshold value, by distance weighted be straight line by the fitting a straight line in every cluster.
9. the Safe belt detection method based on monitoring image according to claim 7, it is characterized in that, the described each pixel calculated respectively on described straight line set cathetus in the horizontal direction or vertical direction before described straight line each side chooses the step of variance yields T of the quadrilateral area that m pixel is formed, also include:
Straight line in described straight line set carries out drift correct.
10. the Safe belt detection method based on video image according to claim 7, it is characterised in that described judge that whether V is more than T, if, it is determined that described straight line is the step of the straight line corresponding with described seat belt image border, particularly as follows:
Judge that whether V is more than T, if, then by described straight line in the horizontal direction or vertical direction n pixel of translation, each pixel calculated after translating on described straight line in the horizontal direction or vertical direction each side choose the variance yields W of the quadrilateral area that n pixel is formed at described straight line；
Judge that whether W is more than T, if, it is determined that described straight line is the straight line that described seat belt image border is corresponding.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410855403.4A CN105809099B (en) | 2014-12-31 | 2014-12-31 | Safety belt detection method based on monitoring image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410855403.4A CN105809099B (en) | 2014-12-31 | 2014-12-31 | Safety belt detection method based on monitoring image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105809099A true CN105809099A (en) | 2016-07-27 |
| CN105809099B CN105809099B (en) | 2019-12-13 |
Family
ID=56464934
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410855403.4A Active CN105809099B (en) | 2014-12-31 | 2014-12-31 | Safety belt detection method based on monitoring image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105809099B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709443A (en) * | 2016-12-19 | 2017-05-24 | 同观科技（深圳）有限公司 | Method for detecting wearing state of safety belt and terminal |
| CN108399357A (en) * | 2017-02-08 | 2018-08-14 | 浙江宇视科技有限公司 | A kind of Face detection method and device |
| CN112232136A (en) * | 2020-09-22 | 2021-01-15 | 北京紫光展锐通信技术有限公司 | Vehicle safety belt detection method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101066081B1 (en) * | 2011-04-04 | 2011-09-20 | (주)에코맥스 | Smart information detection system mounted on the vehicle and smart information detection method using the same |
| CN103150556A (en) * | 2013-02-20 | 2013-06-12 | 西安理工大学 | Safety belt automatic detection method for monitoring road traffic |
| CN103268468A (en) * | 2012-07-06 | 2013-08-28 | 华南理工大学 | Automatic detection method for fastening of safety belts by front sitting persons on motor vehicle |
| CN103522982A (en) * | 2013-10-25 | 2014-01-22 | 公安部第三研究所 | Vehicle safety belt detection method and device based on image analysis |
| CN103870806A (en) * | 2014-02-21 | 2014-06-18 | 杭州奥视图像技术有限公司 | Safety belt detection method combining with steering wheel detection |
-
2014
- 2014-12-31 CN CN201410855403.4A patent/CN105809099B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101066081B1 (en) * | 2011-04-04 | 2011-09-20 | (주)에코맥스 | Smart information detection system mounted on the vehicle and smart information detection method using the same |
| CN103268468A (en) * | 2012-07-06 | 2013-08-28 | 华南理工大学 | Automatic detection method for fastening of safety belts by front sitting persons on motor vehicle |
| CN103150556A (en) * | 2013-02-20 | 2013-06-12 | 西安理工大学 | Safety belt automatic detection method for monitoring road traffic |
| CN103522982A (en) * | 2013-10-25 | 2014-01-22 | 公安部第三研究所 | Vehicle safety belt detection method and device based on image analysis |
| CN103870806A (en) * | 2014-02-21 | 2014-06-18 | 杭州奥视图像技术有限公司 | Safety belt detection method combining with steering wheel detection |
Non-Patent Citations (1)
| Title |
|---|
| 吴法: "图像处理与机器学习在未系安全带驾车检测中的应用", 《博士学位论文全文数据库》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709443A (en) * | 2016-12-19 | 2017-05-24 | 同观科技（深圳）有限公司 | Method for detecting wearing state of safety belt and terminal |
| CN106709443B (en) * | 2016-12-19 | 2020-06-02 | 同观科技（深圳）有限公司 | Method and terminal for detecting wearing state of safety belt |
| CN108399357A (en) * | 2017-02-08 | 2018-08-14 | 浙江宇视科技有限公司 | A kind of Face detection method and device |
| CN108399357B (en) * | 2017-02-08 | 2020-12-29 | 浙江宇视科技有限公司 | Face positioning method and device |
| CN112232136A (en) * | 2020-09-22 | 2021-01-15 | 北京紫光展锐通信技术有限公司 | Vehicle safety belt detection method and device, electronic equipment and storage medium |
| CN112232136B (en) * | 2020-09-22 | 2023-05-05 | 北京紫光展锐通信技术有限公司 | Vehicle safety belt detection method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105809099B (en) | 2019-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102227843B1 (en) | Operating method of lane departure warning system | |
| CN111212772B (en) | Method and device for determining a driving strategy of a vehicle | |
| CN108528336B (en) | Vehicle line pressing early warning system | |
| CN101030256B (en) | Method and apparatus for cutting vehicle image | |
| CA2885019C (en) | Robust windshield detection via landmark localization | |
| US10360460B2 (en) | Method and device of lane departure warning and automobile driving assistance system | |
| WO2012160251A2 (en) | A system, a detection unit, a method and a computer program for detecting passengers' seat belt use in a vehicle | |
| CN103870806B (en) | Safety belt detection method combining with steering wheel detection | |
| CN102785661A (en) | Lane departure control system and lane departure control method | |
| KR101537961B1 (en) | Apparatus and method for detecting seat belt | |
| Guo et al. | Image-based seat belt detection | |
| CN103927548B (en) | Novel vehicle collision avoiding brake behavior detection method | |
| EP2492868A1 (en) | Far-infrared pedestrian detection device | |
| CN105303157A (en) | Algorithm to extend detecting range for AVM stop line detection | |
| KR20110001425A (en) | Lane classification method using statistical model of hsi color information | |
| CN102938057A (en) | Vehicle shadow eliminating method and device | |
| CN105809099A (en) | Safety belt detection method based on monitor image | |
| CN104112141A (en) | Method for detecting lorry safety belt hanging state based on road monitoring equipment | |
| CN106022268A (en) | Identification method and device of speed limiting sign | |
| JP6903477B2 (en) | Orbit identification device and program | |
| Devane et al. | Lane detection techniques using image processing | |
| KR20120086577A (en) | Apparatus And Method Detecting Side Vehicle Using Camera | |
| CN109145797B (en) | Square induction method for road rescue equipment based on vehicle bottom shadow positioning license plate | |
| CN108108656B (en) | Vehicle window corner detection and multidirectional projection-based vehicle window accurate positioning method | |
| CN106803064B (en) | Traffic light rapid identification method |
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 |