CN108198427A - Green light of rushing based on characteristics of image frame is broken rules and regulations determination method - Google Patents
Green light of rushing based on characteristics of image frame is broken rules and regulations determination method Download PDFInfo
- Publication number
- CN108198427A CN108198427A CN201711242925.7A CN201711242925A CN108198427A CN 108198427 A CN108198427 A CN 108198427A CN 201711242925 A CN201711242925 A CN 201711242925A CN 108198427 A CN108198427 A CN 108198427A
- Authority
- CN
- China
- Prior art keywords
- frame
- image
- green light
- image frame
- rushing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of green light of rushing based on characteristics of image frame and breaks rules and regulations determination method, includes the following steps:Step 1, the live video stream for capturing equipment in front end is obtained based on B/S frameworks;Step 2, VideoStream To IMG functions are passed to using the video flowing address of the live video stream as parameter, obtain characteristics of image frame stream;Step 3, picture background modeling is carried out by Gaussian Background model;Step 4, the second frame of extraction characteristics of image frame stream, last frame and intermediate frame, and grey level histogram method is used to judge the similarity of three frame image features frames, export similarity value;Step 5, current traffic state at road cross is judged according to similarity value;Step 6, signal lamp state in characteristics of image frame is identified using deep learning algorithm;Step 7, judge to whether there is in characteristics of image frame stream according to signal lamp state and current traffic state at road cross and rush green light phenomenon;Step 8, it will determine that result and intermediate frame image characteristic frame upload to Snapshot System of Traffic Violation.
Description
Technical field
The present invention relates to image analysis technology field, specifically, relate to a kind of rush green light based on characteristics of image frame
Determination method violating the regulations.
Background technology
As vehicle becomes increasingly popular, traffic congestion and traffic accident annoying urban transportation always, almost per minute to have
Traffic accident is happened at the every nook and cranny in city, and traffic police department increased traffic specification dynamics, while almost each crossing in recent years
There is candid photograph equipment violating the regulations, the behavior of numerous car owners has obtained specification.But with the continuous development of society, caused by rushing green light
Traffic accident rate is riseing year by year, rushes green light and is:Obviously see that junction ahead has resulted in congestion, still crossed after green light
Stop line drives towards intersection, the phenomenon that causing more congestion or even traffic accident.It is difficult sometimes to distinguish due to complicated condition
Be normally travel or, rush green light behavior, therefore solve the problems, such as that judgement rushes green light and has important meaning to the improvement of urban transportation
Justice.
In order to solve the problems, such as present on, people are seeking a kind of ideal technical solution always.
Invention content
To achieve these goals, the technical solution adopted in the present invention is:It is a kind of that green light is rushed based on characteristics of image frame
Determination method violating the regulations, includes the following steps:
Step 1, the live video stream for capturing equipment in front end is obtained based on B/S frameworks;
Step 2, the video flowing address of the live video stream is passed to VideoStreamToIMG functions as parameter, obtained
Take characteristics of image frame stream;
Step 3, picture background modeling is carried out by Gaussian Background model;
Step 4, the second frame of extraction characteristics of image frame stream, last frame and intermediate frame, and judge three using grey level histogram method
The similarity of frame image features frame exports similarity value;
Step 5, current traffic state at road cross is judged according to similarity value;
Step 6, signal lamp state in characteristics of image frame is identified using deep learning algorithm, is green light in signal lamp state
When represented with L=G;
Step 7, the current traffic state at road cross judgement figure obtained according to the signal lamp state and step 5 obtained in step 6
Green light phenomenon is rushed as whether there is in characteristic frame stream;
Step 8, it will determine that result and intermediate frame image characteristic frame upload to Snapshot System of Traffic Violation.
Based on above-mentioned, step 1, based on B/S frameworks obtain front end capture equipment live video stream the specific steps are:
Under B/S frameworks, the SDK interfaces that equipment is captured based on front end are called by video dedicated network, front end is obtained and grabs
The live video stream of equipment is clapped, and preserves video flowing address StreamAd.
Based on above-mentioned, step 4 the specific steps are:
Extract the second frame image IMGArr [2], end-frame image IMGArr [i] and the intermediate frame image of characteristics of image frame
IMGArr[i/2];
Using grey level histogram method to the second frame image IMGArr [2], the end-frame image IMGArr [i] of characteristics of image frame and
Intermediate frame image IMGArr [i/2] carries out similarity-rough set, output similarity value F (G, S, N);
Wherein G, S are grey level histogram, and N is color space number of samples.
Based on above-mentioned, step 5 the specific steps are:
Variable V is enabled to represent current crossing state, current crossing state is judged according to the value of F (G, S, N):
If F (G, S, N)≤70, then V=congestions;
If F (G, S, N)<70 and F (G, S, N)>30, then V=low running speeds;
If F (G, S, N)≤30, then V=is unimpeded.
Based on above-mentioned, the basis for estimation that green light is rushed in step 7 is:
As V=congestions and L=G, then it is judged as rushing green light;Otherwise it is judged as not rushing green light.
The present invention has prominent substantive distinguishing features and significant progress compared with the prior art, and specifically, the present invention is real
When obtain front end capture equipment video flowing, and pass through Gaussian Background model method carry out picture background modeling, from video flowing
It obtains characteristics of image frame and the state at current crossing is had determined that using grey level histogram hair progress similarity judgement, finally combine letter
Signal lamp condition adjudgement has the characteristics that design science, easy to use, can greatly alleviate traffic with the presence or absence of rushing green light phenomenon
Pressure.
Specific embodiment
Below by specific embodiment, technical scheme of the present invention is described in further detail.
The present invention provides a kind of green light of rushing based on characteristics of image frame and breaks rules and regulations determination method, includes the following steps:
Step 1, the live video stream for capturing equipment in front end is obtained based on B/S frameworks;
The SDK interfaces that equipment is captured based on front end are called by video dedicated network, front end is obtained and captures the real-time of equipment
Video flowing, and video flowing address StreamAd is preserved, preserving multiple front ends using arrStreamAd [j] captures the real-time of equipment
Video flowing address;
Step 2, the video flowing address of the live video stream is passed to VideoStreamToIMG functions as parameter, obtained
Characteristics of image frame stream is taken, is represented with array IMGArr [i], wherein i is the index of current frame rate, and characteristics of image frame head frames are
IMGArr [1], the second frame of characteristics of image frame, that is, IMGArr [2], the i-th/2 frame, that is, IMGArr [i/2] of characteristics of image frame, image are special
Levy the i-th frame, that is, IMGArr [i] of frame;
Step 3, picture background modeling is carried out by Gaussian Background model;
Step 4, the second frame of extraction characteristics of image frame stream, last frame and intermediate frame, and judge three using grey level histogram method
The similarity of frame image features frame, output similarity value F (G, S, N):
Wherein G, S are grey level histogram, and N is color space number of samples;
Step 5, current traffic state at road cross is judged according to similarity value;
Variable V is enabled to represent current crossing state, current crossing state is judged according to the value of F (G, S, N):
If F (G, S, N)≤70, then V=congestions;
If F (G, S, N)<70 and F (G, S, N)>30, then V=low running speeds;
If F (G, S, N)≤30, then V=is unimpeded;
Step 6, signal lamp state in characteristics of image frame is identified using deep learning algorithm, is green light in signal lamp state
When represented with L=G;
Step 7, the current traffic state at road cross judgement figure obtained according to the signal lamp state and step 5 obtained in step 6
As, with the presence or absence of green light phenomenon is rushed, specific basis for estimation is in characteristic frame stream:As V=congestions and L=G, then it is judged as rushing green light;
Otherwise it is judged as not rushing green light;
Step 8, capturing system violating the regulations is called to issue the data upwards transmission port on private network server, will determine that result and position
It puts image and uploads to Snapshot System of Traffic Violation.
Specifically, in step 2 by Gaussian Background model method carry out picture background modeling the specific steps are:
Obtain the pixel value and depth value of each pixel in the first frame image of characteristics of image frame;
According to the pixel value and depth value of each pixel, variance and the expectation of the present image are calculated respectively
Value;According to the variance and desired value of the present image, in the multiple single Gauss models included from preset mixed Gauss model
The single Gauss model of one matching of selection;
According to the variance and desired value of the present image, the variance of the single Gauss model of matching corresponding to each pixel
It is updated with desired value, to complete background modeling.
Finally it should be noted that:The above embodiments are merely illustrative of the technical scheme of the present invention and are not intended to be limiting thereof;To the greatest extent
The present invention is described in detail with reference to preferred embodiments for pipe, those of ordinary skills in the art should understand that:Still
It can modify to the specific embodiment of the present invention or equivalent replacement is carried out to some technical characteristics;Without departing from this hair
The spirit of bright technical solution should all cover in the claimed technical solution range of the present invention.
Claims (5)
1. a kind of green light of rushing based on characteristics of image frame is broken rules and regulations determination method, which is characterized in that is included the following steps:
Step 1, the live video stream for capturing equipment in front end is obtained based on B/S frameworks;
Step 2, VideoStream To IMG functions are passed to using the video flowing address of the live video stream as parameter, obtained
Characteristics of image frame stream;
Step 3, picture background modeling is carried out by Gaussian Background model;
Step 4, the second frame of extraction characteristics of image frame stream, last frame and intermediate frame, and three frame figures are judged using grey level histogram method
As the similarity of characteristic frame, similarity value is exported;
Step 5, current traffic state at road cross is judged according to similarity value;
Step 6, signal lamp state in characteristics of image frame is identified using deep learning algorithm, is used when signal lamp state is green light
L=G is represented;
Step 7, the current traffic state at road cross obtained according to the signal lamp state and step 5 obtained in step 6 judges image spy
It whether there is in sign frame stream and rush green light phenomenon;
Step 8, it will determine that result and intermediate frame image characteristic frame upload to Snapshot System of Traffic Violation.
2. a kind of green light of rushing based on characteristics of image frame according to claim 1 is broken rules and regulations determination method, which is characterized in that step
Rapid 1, based on B/S frameworks obtain front end capture equipment live video stream the specific steps are:
Under B/S frameworks, the SDK interfaces that equipment is captured based on front end are called by video dedicated network, front end candid photograph is obtained and sets
Standby live video stream, and preserve video flowing address StreamAd.
3. a kind of green light of rushing based on characteristics of image frame according to claim 1 is broken rules and regulations determination method, it is characterised in that:Step
Rapid 4 the specific steps are:
Extract the second frame image IMGArr [2], end-frame image IMGArr [i] and the intermediate frame image IMGArr of characteristics of image frame
[i/2];
Using grey level histogram method to the second frame image IMGArr [2], end-frame image IMGArr [i] and the centre of characteristics of image frame
Frame image IMGArr [i/2] carries out similarity-rough set, output similarity value F (G, S, N):
Wherein G, S are grey level histogram, and N is color space number of samples.
4. a kind of green light of rushing based on characteristics of image frame according to claim 3 is broken rules and regulations determination method, the specific step of step 5
Suddenly it is:
Variable V is enabled to represent current crossing state, current crossing state is judged according to the value of F (G, S, N):
If F (G, S, N)≤70, then V=congestions;
If F (G, S, N)<70 and F (G, S, N)>30, then V=low running speeds;
If F (G, S, N)≤30, then V=is unimpeded.
5. a kind of green light of rushing based on characteristics of image frame according to claim 4 is broken rules and regulations determination method, green light is rushed in step 7
Basis for estimation be:
As V=congestions and L=G, then it is judged as rushing green light;Otherwise it is judged as not rushing green light.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711242925.7A CN108198427A (en) | 2017-11-30 | 2017-11-30 | Green light of rushing based on characteristics of image frame is broken rules and regulations determination method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711242925.7A CN108198427A (en) | 2017-11-30 | 2017-11-30 | Green light of rushing based on characteristics of image frame is broken rules and regulations determination method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108198427A true CN108198427A (en) | 2018-06-22 |
Family
ID=62573575
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711242925.7A Pending CN108198427A (en) | 2017-11-30 | 2017-11-30 | Green light of rushing based on characteristics of image frame is broken rules and regulations determination method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108198427A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020054210A1 (en) * | 1997-04-14 | 2002-05-09 | Nestor Traffic Systems, Inc. | Method and apparatus for traffic light violation prediction and control |
CN102039738A (en) * | 2009-12-09 | 2011-05-04 | 辉县市文教印务有限公司 | Page online fuzzy identification system of high-speed binding machine |
CN102768801A (en) * | 2012-07-12 | 2012-11-07 | 复旦大学 | Method for detecting motor vehicle green light follow-up traffic violation based on video |
CN103279765A (en) * | 2013-06-14 | 2013-09-04 | 重庆大学 | Steel wire rope surface damage detection method based on image matching |
-
2017
- 2017-11-30 CN CN201711242925.7A patent/CN108198427A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020054210A1 (en) * | 1997-04-14 | 2002-05-09 | Nestor Traffic Systems, Inc. | Method and apparatus for traffic light violation prediction and control |
CN102039738A (en) * | 2009-12-09 | 2011-05-04 | 辉县市文教印务有限公司 | Page online fuzzy identification system of high-speed binding machine |
CN102768801A (en) * | 2012-07-12 | 2012-11-07 | 复旦大学 | Method for detecting motor vehicle green light follow-up traffic violation based on video |
CN103279765A (en) * | 2013-06-14 | 2013-09-04 | 重庆大学 | Steel wire rope surface damage detection method based on image matching |
Non-Patent Citations (1)
Title |
---|
卢志茂等: "基于特征帧构建的运动目标检测方法", 《红外与激光工程》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019105342A1 (en) | Method and apparatus for detecting fake license plates of vehicles, readable storage medium, and electronic device | |
CN103235938B (en) | The method and system of car plate detection and indentification | |
CN103578119B (en) | Target detection method in Codebook dynamic scene based on superpixels | |
CN110795595B (en) | Video structured storage method, device, equipment and medium based on edge calculation | |
WO2019114036A1 (en) | Face detection method and device, computer device, and computer readable storage medium | |
CN109345547A (en) | Traffic lane line detecting method and device based on deep learning multitask network | |
CN109559302A (en) | Pipe video defect inspection method based on convolutional neural networks | |
CN108681692A (en) | Increase Building recognition method in a kind of remote sensing images based on deep learning newly | |
CN104050471A (en) | Natural scene character detection method and system | |
CN108921083A (en) | Illegal flowing street pedlar recognition methods based on deep learning target detection | |
CN103400113B (en) | Freeway tunnel pedestrian detection method based on image procossing | |
CN111047874B (en) | Intelligent traffic violation management method and related product | |
CN112464822B (en) | Helmet wearing detection method and system based on feature enhancement | |
WO2019119515A1 (en) | Face analysis and filtering method, device, embedded apparatus, dielectric and integrated circuit | |
CN110097026A (en) | A kind of paragraph correlation rule evaluation method based on multidimensional element Video segmentation | |
CN107564018A (en) | It is a kind of to utilize the method for improving iterative algorithm extraction target image | |
CN107066963A (en) | A kind of adaptive people counting method | |
CN110135345A (en) | Activity recognition method, apparatus, equipment and storage medium based on deep learning | |
CN106886763A (en) | The system and method for real-time detection face | |
CN108734131A (en) | A kind of traffic sign symmetry detection methods in image | |
US20230386221A1 (en) | Method for detecting road conditions and electronic device | |
Shan et al. | A small traffic sign detection algorithm based on modified ssd | |
CN110796580B (en) | Intelligent traffic system management method and related products | |
CN109784214A (en) | A kind of detection device and method of railroad track foreign matter | |
Tariq et al. | Real time vehicle detection and colour recognition using tuned features of Faster-RCNN |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180622 |
|
RJ01 | Rejection of invention patent application after publication |