CN105809099B - Safety belt detection method based on monitoring image - Google Patents
Safety belt detection method based on monitoring image Download PDFInfo
- Publication number
- CN105809099B CN105809099B CN201410855403.4A CN201410855403A CN105809099B CN 105809099 B CN105809099 B CN 105809099B CN 201410855403 A CN201410855403 A CN 201410855403A CN 105809099 B CN105809099 B CN 105809099B
- Authority
- CN
- China
- Prior art keywords
- image
- straight line
- safety belt
- pixels
- monitoring
- 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
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
the invention provides a safety belt detection method based on a monitoring image, which comprises the following steps: acquiring a monitoring image to be detected; identifying a license plate in the image, and determining a position area of a vehicle driver according to the position of the license plate; intercepting an image of the position area of the driver; and detecting whether the intercepted image contains a safety belt image. The invention carries out image recognition analysis on the vehicle driver area in the monitoring video image, reduces the range of image recognition analysis, can effectively improve the detection efficiency of the wearing state of the safety belt, simultaneously, adopts the method of detecting the straight line corresponding to the safety belt edge image after carrying out gray processing on the video monitoring image, reduces the image quality requirement on the monitoring video image, can carry out accurate detection on various monitoring images with different pixel sizes and definitions, and improves the general applicability of the scheme.
Description
Technical Field
The invention relates to the technical field of image recognition and pattern recognition, in particular to a safety belt detection method based on a monitoring image.
background
Along with the rapid growth of social economy, the material living standard of people is continuously improved, meanwhile, the automobile industry develops at a high speed, more and more families select automobiles as transportation means for travel and travel, and the modern travel mode provides comfortable and convenient travel selection for people and also has accidents and potential safety hazards brought by the fact that drivers or passengers violate road traffic regulations.
Whether the safety belt is worn or not in the driving process of the vehicle directly relates to the life safety of a driver and passengers, and when collision happens, the safety belt can fix the driver and passengers in a certain space range around the vehicle seat, so that serious injury caused by collision between the body out of control and objects in the vehicle or throwing of the object out of the vehicle is avoided. The current road traffic safety regulations in China are clearly stipulated, and when a motor vehicle runs, a driver and passengers should use safety belts according to the regulations.
At present, road traffic monitoring systems are increasingly popularized and improved, vehicles running on roads can be accurately shot to obtain clear video images, and necessary basic conditions are provided for confirming illegal behaviors that drivers and passengers do not wear safety belts according to regulations in the process of running of the vehicles through the road traffic monitoring systems. At present, the mode of manually detecting shot video images is mainly adopted to judge road traffic illegal behaviors, and the data volume of video monitoring images is huge, so that the working efficiency of manual operation is low, and the accuracy is not high. Therefore, how to automatically, efficiently and intelligently identify and process the video monitoring image and quickly and accurately detect the violation without wearing the safety belt becomes a technical problem to be solved urgently.
Disclosure of Invention
In order to solve at least one of the above technical problems, an object of the present invention is to provide a method for detecting a seat belt based on a monitored image, which can accurately detect and determine whether a driver of a vehicle wears the seat belt by performing recognition and analysis on the monitored image.
In view of this, the present invention provides a safety belt detection method based on a monitoring image, including: acquiring a monitoring image to be detected; identifying a license plate in the image, and determining a position area of a vehicle driver according to the position of the license plate; intercepting an image of the position area of the driver; and detecting whether the intercepted image contains a safety belt image.
Preferably, the step of identifying the license plate in the image and determining the position area of the vehicle driver according to the position of the license plate is specifically to identify the license plate in the image, and select an area with a preset width and a preset height as the position area of the vehicle driver by taking the position of the upper right corner of the image area of the license plate as an end point.
preferably, before the step of detecting whether the intercepted image includes a seat belt image, the method further includes performing face detection on the intercepted image of the position area of the driver, and intercepting an image of the upper torso area of the driver according to the detected face position.
preferably, the step of detecting whether the intercepted image includes a seat belt image is performed, specifically, image enhancement and edge extraction are performed on the intercepted image, and a straight line set meeting a preset condition in the image is obtained; and detecting straight lines in the straight line set, and determining whether a straight line corresponding to the edge of the safety belt image exists.
Preferably, the step of performing image enhancement and edge extraction on the captured image to obtain a straight line set meeting a preset condition in the image is specifically to perform gray processing, image filtering, edge detection and/or noise filtering on the captured image, and detect and obtain a straight line set meeting a preset slope in the captured image.
preferably, the straight line in the set of straight lines comprises at least 10 pixels, and the slope of the straight line is greater than 0.5 and less than 3.7.
Preferably, the step of detecting the straight lines in the set and determining whether there is a straight line corresponding to the edge of the image of the seat belt is performed by calculating a variance value T of a quadrilateral region formed by selecting m pixels on each of the left and right sides of each of the pixels on the straight line in the straight line set along a horizontal direction or a vertical direction; calculating a variance value L of a quadrilateral region formed by selecting n pixels on the left side of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line, and calculating a variance value R of a quadrilateral region formed by selecting n pixels on the right side of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line, wherein n is greater than m; respectively judging whether L is smaller than T and whether R is smaller than T, if L is smaller than T and/or R is smaller than T, calculating a variance value V of a quadrilateral area formed by respectively selecting n pixels on the left side and the right side of the straight line along the horizontal direction or the vertical direction of each pixel on the straight line; and judging whether V is larger than T, if so, determining the straight line as the straight line corresponding to the edge of the safety belt image.
Preferably, before the step of calculating the variance value T of the quadrilateral region formed by selecting m pixels on each of the left and right sides of the straight line along the horizontal direction or the vertical direction of each pixel on the straight line in the straight line set, the method further includes selecting all parallel straight lines in the straight line set, dividing every two straight lines with a distance smaller than a preset threshold value in the straight line set into a cluster, and fitting the straight lines in each cluster into a straight line through distance weighting.
Preferably, before the step of calculating the variance value T of the quadrilateral region formed by selecting m pixels on the left side and the right side of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line in the straight line set, the method further includes performing drift correction on the straight line in the straight line set.
preferably, the step of determining whether V is greater than T, and if so, that the straight line is the straight line corresponding to the edge of the seat belt image specifically includes: judging whether V is larger than T, if so, translating the straight line by n pixels along the horizontal direction or the vertical direction, and calculating the variance value W of a quadrilateral area formed by respectively selecting n pixels at the left side and the right side of the straight line along the horizontal direction or the vertical direction of each pixel on the translated straight line; and judging whether W is larger than T, if so, determining the straight line as the straight line corresponding to the edge of the safety belt image.
The invention provides a safety belt detection method based on a monitoring image, which is characterized in that an image area of a vehicle driver is determined according to the position of a license plate in a road traffic monitoring video image, the image of the area is processed to obtain a straight line set which accords with a preset condition in the image, whether a straight line corresponding to an edge image of a safety belt exists or not is determined by traversing the straight line set, and if the straight line set exists, the vehicle driver in the image is determined to wear the safety belt. The invention carries out image recognition analysis on the vehicle driver area in the monitoring video image, reduces the range of image recognition analysis, can effectively improve the detection efficiency of the wearing state of the safety belt, simultaneously, adopts the method of detecting the straight line corresponding to the safety belt edge image after carrying out gray processing on the video monitoring image, reduces the image quality requirement on the monitoring video image, can carry out accurate detection on various monitoring images with different pixel sizes and definitions, and improves the general applicability of the scheme.
Drawings
in order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts:
Fig. 1 shows a flow chart of a seat belt detection method based on a monitoring image according to an embodiment of the invention.
Detailed Description
So that the objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof that are illustrated in the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, this is merely an example of the present invention, which may be embodied in other ways than is specifically described herein, and therefore the scope of the present invention is not limited by the specific examples disclosed below.
fig. 1 shows a flow chart of a seat belt detection method based on a monitoring image according to an embodiment of the invention.
The safety belt detection method based on the monitoring image comprises the following steps: step 101, acquiring a monitoring image to be detected; step 102, identifying a license plate in the image, and determining a position area of a driver of the vehicle according to the position of the license plate; step 103, intercepting an image of a position area of a driver; 104, performing image enhancement and edge processing on the intercepted image to obtain a straight line set which meets a preset condition in the image; and 105, detecting straight lines in the straight line set, and determining whether straight lines corresponding to the edges of the safety belt images exist.
In the technical scheme, a monitoring video image to be detected is obtained, wherein the monitoring video image comprises a front image of a front part of a vehicle, the position of a license plate is determined by recognizing the license plate in the image, and the position area of a driver of the vehicle is determined according to the position and the size ratio of the license plate.
In the above technical solution, preferably, in step 104, the intercepted image of the driver position area is subjected to face detection, the seat belt detection area is analyzed and positioned according to the detected face position, the image of the seat belt detection area is further intercepted, image enhancement and edge processing are performed on the intercepted image, and a straight line set L1 meeting the preset condition in the image is obtained.
In the technical scheme, the intercepted image of the position area of the driver is subjected to face detection, the image area of the upper body trunk of the vehicle driver is analyzed and determined according to the detected face position area, the image of the area is intercepted, the intercepted image is subjected to image enhancement and edge processing, and a straight line set which meets preset conditions in the image is obtained, so that the image processing range is further reduced, and the image processing efficiency is improved.
In the foregoing technical solution, preferably, in step 104, the captured image is subjected to gray processing, image filtering, edge detection and/or noise filtering, so as to obtain a straight line set meeting a preset condition in the image. Specifically, the method comprises the following steps: the image edge detection method comprises the steps of carrying out image edge detection on an intercepted image by adopting a Canny operator, detecting and extracting a straight Line set L1 which accords with a preset slope in the intercepted image through an LSD (Line Segment Detector) straight Line detection algorithm, wherein a straight Line in the straight Line set comprises at least 10 pixels, and the slope of the straight Line is more than 0.5 and less than 3.7.
In the technical scheme, according to the complexity of the video monitoring image, a plurality of straight lines with different lengths and close distances may exist in the straight line set obtained in step 104, and step 104 further includes further screening and optimizing the straight line set. Specifically, the method comprises the following steps: dividing straight lines which are parallel to each other and have a distance of less than 5 pixels in the straight line set L1 into a cluster, weighting and fitting all the straight lines contained in the cluster and the cluster center distance by an average method or a least square method to obtain a new straight line, and replacing all original straight lines in the cluster with the straight line to obtain an optimized straight line set L2.
In the above technical solution, a straight line that meets a preset slope in an image detected and extracted by an LSD straight line detection algorithm may have a certain degree of pixel deviation, that is, a start point and an end point of the straight line are not completely located in a boundary region, but deviate by a plurality of pixels within a certain range of the boundary region. The step 104 further comprises performing a drift correction on the straight lines in the set. Specifically, the method comprises the following steps: respectively selecting K pixels on the left and right of the straight line between the starting point and the end point of the straight line along the direction vertical to the straight line, setting the arrangement of the K pixels to accord with Gaussian distribution, calculating a local maximum value as a new starting point and a new local point of the straight line, and accurately distributing all the straight lines in the L2 set in a local boundary region, thereby obtaining a corrected straight line set L3.
in the above technical solution, in the step 105, variance values T of quadrilateral areas formed by selecting m pixels on the left and right sides of the straight line along the horizontal direction or the vertical direction respectively at each pixel between the start point and the end point of the straight line a1 in the straight line set L3 are respectively calculated, where m is 3. Then, calculating a variance value L of a quadrilateral region formed by selecting n pixels on the left side of the straight line A1 along the horizontal direction or the vertical direction of each pixel on the straight line A1, and calculating a variance value R of a quadrilateral region formed by selecting n pixels on the right side of the straight line A1 along the horizontal direction or the vertical direction of each pixel on the straight line A1, if L is smaller than T, determining that the left region of the straight line A1 is a safety belt image possible region, and setting a Mark _1 for the straight line A1; if R is less than T, determining the right area of the straight line A1 as the potential area of the safety belt image, and setting a Mark Mark _2 for the straight line A1; if L is smaller than T and R is smaller than T, determining the left and right areas of the straight line A1 as the possible areas of the safety belt image, and setting a Mark Mark _3 for the straight line A1; and if both L and R are larger than T, determining that the left side and the right side of the straight line A1 are not the possible areas of the safety belt image, and setting a Mark Mark _0 for the straight line.
In the technical scheme, after traversing all straight lines in a straight line set L3, all straight lines which are not marked as Mark _0 in the straight line set L3 are selected, a variance value V of a quadrilateral region formed by respectively selecting n pixels on the left side and the right side of the straight line along the horizontal direction or the vertical direction by taking each pixel on the straight line as the center is calculated, and if V is larger than T, the straight line is determined to be a straight line corresponding to the edge of a safety belt image, namely the safety belt image exists in the intercepted video image.
In the above technical solution, in order to further ensure the accuracy of detecting the image of the seat belt, when it is determined that V is greater than T, the straight line is translated by n pixels in the horizontal direction or the vertical direction, and specifically, if the Mark of the straight line is Mark _1, the straight line is translated by n pixels to the left of the straight line in the horizontal direction or the vertical direction; if the Mark of the straight line is Mark _1, translating the straight line to the right side of the straight line by n pixels along the horizontal or vertical direction; and if the Mark of the straight line is Mark _3, respectively translating the straight line to the left side and the right side of the straight line by n pixels along the horizontal direction or the vertical direction. And calculating a variance value W of a quadrilateral area formed by respectively selecting n pixels on the left side and the right side of the line along the horizontal direction or the vertical direction of each pixel on the translated line, judging whether the W is greater than the T, and if so, determining the line as the line corresponding to the edge of the security image.
In the technical scheme, the calculated straight line corresponding to the edge of the safety belt image is translated to the area of the safety belt image, the variance value of the area on the two sides of the straight line is calculated by taking the translated straight line as the center, and the straight line corresponding to the edge of the safety belt image is further determined by comparison and judgment, so that the accuracy and the stability of the detection result are ensured.
The safety belt detection method based on the monitoring image comprises the steps of intercepting an image of a vehicle driver area in a road monitoring video image, extracting a straight line set which meets the conditions of preset length and slope in the image through image gray processing, noise filtering and the like, detecting straight lines in the straight line set to determine whether straight lines corresponding to the edge of a safety belt image exist or not, and determining that the vehicle driver wears the safety belt in the image if the straight lines exist. The invention intercepts the image area of the vehicle driver for detection by identifying the characteristic area of the monitoring image, can quickly and accurately detect the wearing condition of the safety belt of the vehicle driver, and effectively improves the efficiency and the stability of the detection of the urban road traffic safety monitoring image.
It is again stated that all of the features disclosed in this specification, or all of the steps in any method or process so disclosed, may be combined in any combination, except mutually exclusive features and/or steps.
Any feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving equivalent or similar purposes, unless expressly stated otherwise. That is, unless expressly stated otherwise, each feature is only an example of a generic series of equivalent or similar features.
the invention is not limited to the foregoing embodiments. The invention extends to any novel feature or any novel combination of features disclosed in this specification, and to any novel method or process steps or any novel combination of features disclosed.
Claims (9)
1. A safety belt detection method based on monitoring images is characterized by comprising the following steps:
Acquiring a monitoring image to be detected;
Identifying a license plate in the image, and determining a position area of a vehicle driver according to the position of the license plate;
Intercepting an image of the position area of the driver;
Carrying out image enhancement and edge extraction on the intercepted image to obtain a straight line set which accords with a preset condition in the image;
respectively calculating the variance value T of a quadrilateral area formed by selecting m pixels on the left side and the right side of the straight line along the horizontal direction or the vertical direction of each pixel on the straight line in the straight line set;
Calculating a variance value L of a quadrilateral region formed by selecting n pixels on the left side of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line, and calculating a variance value R of a quadrilateral region formed by selecting n pixels on the right side of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line, wherein n is greater than m;
Respectively judging whether L is smaller than T and whether R is smaller than T, if L is smaller than T or R is smaller than T, calculating a variance value V of a quadrilateral area formed by respectively selecting n pixels on the left side and the right side of the straight line along the horizontal direction or the vertical direction of each pixel on the straight line;
And judging whether V is larger than T, if so, determining the straight line as the straight line corresponding to the edge of the safety belt image.
2. the monitoring-image-based safety belt detection method according to claim 1, wherein the step of identifying the license plate in the image and determining the position area of the vehicle driver according to the position of the license plate comprises:
And recognizing the license plate in the image, taking the upper right corner position of the license plate image area as an end point, and selecting an area with a preset width and a preset height as a position area of a vehicle driver.
3. The method for detecting a seat belt based on a monitoring image according to claim 2, wherein before the step of performing image enhancement and edge extraction on the intercepted image to obtain a set of straight lines meeting a preset condition in the image, the method further comprises:
And carrying out face detection on the intercepted image of the position area of the driver, and intercepting the image of the upper body trunk area of the driver according to the detected face position.
4. The safety belt detection method based on the monitoring image according to any one of claims 1 to 3, wherein the step of performing image enhancement and edge extraction on the intercepted image to obtain a straight line set meeting a preset condition in the image specifically comprises:
and carrying out gray processing, image filtering, edge detection and/or noise filtering processing on the intercepted image, and detecting and acquiring a straight line set which accords with a preset slope in the intercepted image.
5. The surveillance image-based seat belt detection method according to claim 4, wherein a straight line in the set of straight lines includes at least 10 pixels, and a slope of the straight line is greater than 0.5 and less than 3.7.
6. The method for detecting a seat belt based on a monitored image according to claim 5, wherein after the steps of determining whether L is less than T and R is less than T respectively, further comprising:
If L is smaller than T and R is smaller than T, calculating a variance value V of a quadrilateral area formed by respectively selecting n pixels on the left side and the right side of the straight line along the horizontal direction or the vertical direction of each pixel on the straight line;
And judging whether V is larger than T, if so, determining the straight line as the straight line corresponding to the edge of the safety belt image.
7. The method for detecting a seat belt based on a monitored image according to claim 6, wherein before the step of calculating the variance value T of the quadrilateral region formed by selecting m pixels on the left and right sides of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line in the straight line set, the method further comprises:
And selecting all parallel straight lines in the straight line set, dividing every two straight lines with the distance smaller than a preset threshold value in the straight line set into a cluster, and fitting the straight lines in each cluster into a straight line through distance weighting.
8. the method for detecting a seat belt based on a monitored image according to claim 6, wherein before the step of calculating the variance value T of the quadrilateral region formed by selecting m pixels on the left and right sides of the straight line along the horizontal direction or the vertical direction for each pixel on the straight line in the straight line set, the method further comprises:
and performing drift correction on the straight lines in the straight line set.
9. The method for detecting a seat belt based on a monitored image according to claim 6, wherein the step of determining whether V is greater than T, if so, determining that the straight line is a straight line corresponding to the edge of the seat belt image is specifically as follows:
Judging whether V is larger than T, if so, translating the straight line by n pixels along the horizontal direction or the vertical direction, and calculating the variance value W of a quadrilateral area formed by respectively selecting n pixels at the left side and the right side of the straight line along the horizontal direction or the vertical direction of each pixel on the translated straight line;
and judging whether W is larger than T, if so, determining the straight line as the straight line corresponding to the edge of the safety belt image.
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 CN105809099A (en) | 2016-07-27 |
| CN105809099B true 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) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN108399357B (en) * | 2017-02-08 | 2020-12-29 | 浙江宇视科技有限公司 | Face positioning method and device |
| 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 |
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 |
|---|
| 图像处理与机器学习在未系安全带驾车检测中的应用;吴法;《博士学位论文全文数据库》;20140815;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105809099A (en) | 2016-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111212772B (en) | Method and device for determining a driving strategy of a vehicle | |
| CN106326822B (en) | Method and device for detecting lane line | |
| CN103996053B (en) | Lane departure alarm method based on machine vision | |
| US8902053B2 (en) | Method and system for lane departure warning | |
| KR101605514B1 (en) | Apparatus and method for recognizing lane | |
| CN101339616B (en) | Roads recognition method and apparatus | |
| US9257043B2 (en) | Lane correction system, lane correction apparatus and method of correcting lane | |
| WO2012160251A2 (en) | A system, a detection unit, a method and a computer program for detecting passengers' seat belt use in a vehicle | |
| US8599257B2 (en) | Vehicle detection device, vehicle detection method, and vehicle detection program | |
| KR101094752B1 (en) | Lane Classification Method Using Statistical Model of HSI Color Information | |
| Guo et al. | Image-based seat belt detection | |
| CN108528336A (en) | A kind of vehicle crimping gives warning in advance system | |
| US11055542B2 (en) | Crosswalk marking estimating device | |
| JP2010244456A (en) | Boundary line recognition apparatus | |
| JPWO2012023412A1 (en) | Three-dimensional object detection apparatus and three-dimensional object detection method | |
| US20110216944A1 (en) | In-vehicle white line recognition apparatus | |
| JP4744537B2 (en) | Driving lane detector | |
| CN104112141A (en) | Method for detecting lorry safety belt hanging state based on road monitoring equipment | |
| CN103870806B (en) | Safety belt detection method combining with steering wheel detection | |
| CN105809099B (en) | Safety belt detection method based on monitoring image | |
| US20110013839A1 (en) | Procede de determination d'une region d'interet dans une image | |
| CN104123557A (en) | Method for detecting car safety belt fastening state based on road monitoring device | |
| JP5204144B2 (en) | In-vehicle white line recognition device | |
| CN106803064B (en) | Traffic light rapid identification method | |
| CN112989886A (en) | Real-time vehicle line pressing advance early warning 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 |