CN109165676A - A kind of round-the-clock highway fog grade monitoring method based on video analysis - Google Patents
A kind of round-the-clock highway fog grade monitoring method based on video analysis Download PDFInfo
- Publication number
- CN109165676A CN109165676A CN201810845049.5A CN201810845049A CN109165676A CN 109165676 A CN109165676 A CN 109165676A CN 201810845049 A CN201810845049 A CN 201810845049A CN 109165676 A CN109165676 A CN 109165676A
- Authority
- CN
- China
- Prior art keywords
- fog
- grade
- picture
- night
- round
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 19
- 239000003595 mist Substances 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 238000013135 deep learning Methods 0.000 claims abstract description 4
- 230000000694 effects Effects 0.000 claims abstract description 4
- 238000012800 visualization Methods 0.000 claims abstract description 4
- 230000004069 differentiation Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The round-the-clock highway fog grade monitoring method based on video analysis that the invention discloses a kind of, steps are as follows: acquisition monitors the true picture of regional highway, the picture being collected into carries out manual sort, by Darknet deep learning frame respectively to daytime, night sample set carries out classifier training, the classifier obtained using step S3 is to the image obtained from live video stream, the classifier at daytime or night is selected to classify according to the time, to realize round-the-clock highway fog grade monitoring, it will test in result and detection picture address information deposit MySQL database, front-end interface accesses database, by current fog grade and historical evolution, newest monitored picture is shown on map, intuitive effect of visualization is provided, collect detection picture and testing result, such method can Group's mist of locality is provided and is accurately positioned in real time, vehicular traffic can be reminded, speed limit, to reduce accident amount, bring better prospect of the application.
Description
Technical field
The invention belongs to technical field of computer vision, special suitable for the round-the-clock monitoring to highway fog grade
It is not the group's mist phenomenon offer early warning that happens suddenly to highway part way.
Background technique
So far, China's highway mileage open to traffic number occupies the first in the world, highway tool more than 130,000 kilometers
There is the characteristics of speed is fast, vehicle flowrate is big, closing, therefore expressway visibility is to influence the principal element of the person, vehicle safety, according to
Statistics, dense fog (including a mist) are the main inducings for leading to the similar major traffic accidents that shunt into one another, and group's mist is also known as radiation fog, is
Caused by the air mass cooling close to road surface, with locality, sudden, concentration is big, moveable feature, referred to as " high
The flowing killer of fast highway ", since mist range is small, monitoring and forecast difficulty it is larger, more reasonable solution be
More monitoring sites are set on highway, it is early to find, it is early to prevent.
Summary of the invention
The round-the-clock highway fog grade monitoring based on video analysis that the main purpose of the present invention is to provide a kind of
Method can effectively solve the problems in background technique.
To achieve the above object, the technical scheme adopted by the invention is as follows:
A kind of round-the-clock highway fog grade monitoring method based on video analysis, follows the steps below to implement:
Step S1: different periods that acquisition regional freeway surveillance and control camera to be monitored takes, different fog grades it is true
Real picture；
Step S2: the picture that step S1 is collected into carries out manual sort, daytime, night is separated, respectively to the figure on daytime, night
Piece carries out sophisticated category again, is divided into fogless, mist, mist, dense fog, thick fog, strong thick fog, invalid 7 grades；
Step S3: by Darknet deep learning frame respectively to daytime, night sample set training classifier；
Step S4: the classifier obtained using step S3 selects daytime to the image obtained from live video stream, according to the time
Or the classifier at night is classified, thus realize round-the-clock highway fog grade monitoring, in addition, to prevent from judging by accident,
Continuous acquisition two field pictures carry out fog grade judgement respectively, reduce according to certain logic and judge by accident；
Step S5: will test in result and detection picture address information deposit MySQL database, and front-end interface accesses database,
Current fog grade, historical evolution and newest monitored picture are shown on map, intuitive effect of visualization is provided；
Step S6: collecting detection picture and the classification sample of erroneous judgement is particularly added in sample set and holds again by testing result
Row step S3, results model is substituted into step S4, repeats the step as needed, until fog grade testing result reaches
The required accuracy requirement.
2. a kind of round-the-clock highway fog grade monitoring method based on video analysis according to right 1,
It is characterized in that, in the step S1, to guarantee that classifier is adapted to practical highway scene in step S3, specially chooses wait supervise
The scene picture that geodetic area highway actually photographed is as training sample.
3. a kind of round-the-clock highway fog grade monitoring method based on video analysis according to right 1,
It is characterized in that, in the step S2, since daytime, night monitor at a high speed the picture taken and differ greatly, to improve step S3
Samples pictures are divided into daytime, night two parts, carry out classifier training respectively by the classifying quality of middle classifier, according to comparing
Common fog grade classification is classified as fogless, mist, mist, dense fog, thick fog, strong thick fog, particularly, flower is shielded, is exposed,
Night, big car light situation was divided into the 7th invalid class of class.
4. a kind of round-the-clock highway fog grade monitoring method based on video analysis according to right 1,
It is characterized in that, in the step S4, the time division on daytime, night is carried out according to the sunrise in different months, sunset time, in reality
Continuous acquisition two field pictures are used when border is monitored current fog grade, determine the mode of fog grade respectively, separately
It reduces and judges by accident according to following logic:
A1: if grade judgement twice is identical, and being c grades, then current fog grade is c grades；
A2: if grade determines difference twice, and once identical as history grade, then current fog grade is history grade；
A3: if grade determines difference twice, and not identical as history grade, then two field pictures are acquired again and carry out classification grade
Judgement.
5. a kind of round-the-clock highway fog grade monitoring method based on video analysis according to right 1,
It is characterized in that, in the step S6, to improve fog grade monitoring accuracy, periodically collects history grade discrimination picture and differentiate knot
Fruit is added after erroneous judgement through the artificial sample correctly classified, re -training classifier, to improve mist in original sample collection basis
The precision of gas grade monitoring.
Compared with prior art, the invention has the following beneficial effects:
In the present invention, using built video monitoring equipment, it can be realized without additionally increasing front end hardware and automatically analyze height in real time
The fog grade function of fast highway different sections of highway；Can group's mist to locality provide and be accurately positioned in real time, mentioned for traffic police department
For foundation of enforcing the law, meets burst group's mist or dense fog situation can remind vehicular traffic, speed limit subtracts to reduce accident amount
Few casualties and property loss, bring better prospect of the application.
Detailed description of the invention
Fig. 1 is specific flow chart of the present invention；
Fig. 2 is that single fog grade specifically differentiates logical flow chart.
Specific embodiment
In a specific embodiment, it is public that the round-the-clock high speed based on video analysis will be described clear and completely in conjunction with attached drawing
The detailed process of road fog grade monitoring method.It should be pointed out that these descriptions are example, it is not intended to limit the present invention
Range.In addition, in the following description, descriptions of well-known structures and technologies are omitted, this hair is obscured to avoid unnecessary
Bright concept.
Specific step is as follows:
Step 1: region expressway video monitoring distinct device to be monitored, different time sections, the history of different fog grades are acquired
Video clip parses the video into picture one by one；
Step 2: a large amount of pictures that will be parsed in step 1, by manually classifying, respectively by the picture on daytime, night
It is divided into following 7 grades:
Step 3: the classifier on daytime, night sample set is trained respectively by Darknet deep learning frame；
Step 4: handling library by FFmpeg multimedia video and access region highway video monitoring acquisition picture to be monitored,
Classify to picture, to obtain current fog grade, single fog grade specifically differentiates that logic flow is as shown in Figure 2；
Step 5: will test in result and detection picture address information deposit MySQL database, and front-end interface accesses database,
Current fog grade, historical evolution and newest monitored picture are shown on map, intuitive effect of visualization is provided；
Step 6: collecting detection picture and the classification sample of erroneous judgement is particularly added in sample set and holds again by testing result
Row step 3, results model is substituted into step 4, repeats the step as needed, until fog grade monitoring result reaches
The required accuracy requirement.
The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (5)
1. a kind of round-the-clock highway fog grade monitoring method based on video analysis, follows the steps below to implement:
Step S1: different periods that acquisition regional freeway surveillance and control camera to be monitored takes, different fog grades it is true
Real picture；
Step S2: the picture that step S1 is collected into carries out manual sort, daytime, night is separated, respectively to the figure on daytime, night
Piece carries out sophisticated category again, is divided into fogless, mist, mist, dense fog, thick fog, strong thick fog, invalid 7 grades；
Step S3: classifier training is carried out to daytime, night sample set respectively by Darknet deep learning frame；
Step S4: the classifier obtained using step S3 selects daytime to the image obtained from live video stream, according to the time
Or the classifier at night is classified, thus realize round-the-clock highway fog grade monitoring, in addition, to prevent from judging by accident,
Continuous acquisition two field pictures carry out fog grade judgement respectively, reduce according to certain logic and judge by accident；
Step S5: will test in result and detection picture address information deposit MySQL database, and front-end interface accesses database,
By current fog grade and historical evolution, newest monitored picture is shown on map, provides intuitive effect of visualization；
Step S6: collecting detection picture and the classification sample of erroneous judgement is particularly added in sample set and holds again by testing result
Row step S3, results model is substituted into step S4, repeats the step as needed, until fog grade monitoring result reaches
The required accuracy requirement.
2. a kind of round-the-clock highway fog grade monitoring method based on video analysis, feature according to right 1
It is, in the step S1, to guarantee that classifier is adapted to practical highway scene in step S3, specially chooses to be monitoredly
The scene picture that area's highway actually photographed is as training sample.
3. a kind of round-the-clock highway fog grade monitoring method based on video analysis, feature according to right 1
It is, in the step S2, since daytime, night monitor at a high speed the picture taken and differ greatly, divides to improve in step S3
Samples pictures are divided into daytime, night two parts, classifier training are carried out respectively, according to relatively common by the classifying quality of class device
Fog grade classification, be classified as fogless, mist, mist, dense fog, thick fog, strong thick fog, particularly, will flower screen, exposure, night
Big car light situation is divided into the 7th invalid class of class.
4. a kind of round-the-clock highway fog grade monitoring method based on video analysis, feature according to right 1
It is, in the step S4, the time division on daytime, night is carried out according to the sunrise in different months, sunset time, practical right
Using continuous acquisition two field pictures when current fog grade is monitored, determine the mode of fog grade respectively, separately according to
Following logic reduces erroneous judgement:
A1: if grade judgement twice is identical, and being c grades, then current fog grade is c grades；
A2: if grade determines difference twice, and once identical as history grade, then current fog grade is history grade；
A3: if grade determines difference twice, and not identical as history grade, then two field pictures are acquired again and carry out classification grade
Judgement.
5. a kind of round-the-clock highway fog grade monitoring method based on video analysis, feature according to right 1
Be, in the step S6, to improve fog grade monitoring accuracy, periodically collect history grade discrimination picture and differentiation as a result,
It is added in original sample collection basis after erroneous judgement through the artificial sample correctly classified, re -training classifier, to improve fog
The precision of grade monitoring.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810845049.5A CN109165676A (en) | 2018-07-27 | 2018-07-27 | A kind of round-the-clock highway fog grade monitoring method based on video analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810845049.5A CN109165676A (en) | 2018-07-27 | 2018-07-27 | A kind of round-the-clock highway fog grade monitoring method based on video analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109165676A true CN109165676A (en) | 2019-01-08 |
Family
ID=64898535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810845049.5A Pending CN109165676A (en) | 2018-07-27 | 2018-07-27 | A kind of round-the-clock highway fog grade monitoring method based on video analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109165676A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886920A (en) * | 2019-01-16 | 2019-06-14 | 安徽谛听信息科技有限公司 | A kind of greasy weather stage division, greasy weather hierarchy system |
| CN109934780A (en) * | 2019-02-21 | 2019-06-25 | 北京以萨技术股份有限公司 | A kind of traffic surveillance videos defogging method based on dark primary priori |
| CN111598885A (en) * | 2020-05-21 | 2020-08-28 | 公安部交通管理科学研究所 | Automatic visibility grade marking method for highway foggy pictures |
| CN112419745A (en) * | 2020-10-20 | 2021-02-26 | 中电鸿信信息科技有限公司 | Highway group fog early warning system based on degree of depth fusion network |
| CN112749654A (en) * | 2020-12-31 | 2021-05-04 | 南京恩瑞特实业有限公司 | Deep neural network model construction method, system and device for video fog monitoring |
| CN113435405A (en) * | 2021-07-15 | 2021-09-24 | 山东交通学院 | Expressway night fog monitoring method and system based on video images |
| CN115310519A (en) * | 2022-07-11 | 2022-11-08 | 广州国交润万交通信息有限公司 | A method for weather discrimination combining video and audio |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102792322A (en) * | 2009-09-30 | 2012-11-21 | 天宝导航有限公司 | Visual organization of information via associated geospatial data |
| CN103150914A (en) * | 2013-02-04 | 2013-06-12 | 安徽达尔智能控制系统有限公司 | Visual intelligent transportation meteorological monitoring and forewarning management system based on Web geographic information system (GIS) |
| CN103903008A (en) * | 2014-03-26 | 2014-07-02 | 国家电网公司 | Power transmission line fog level recognition method and system based on images |
| CN104463196A (en) * | 2014-11-11 | 2015-03-25 | 中国人民解放军理工大学 | Video-based weather phenomenon recognition method |
| CN106556579A (en) * | 2016-11-07 | 2017-04-05 | 南京理工大学 | Group's mist image all-weather self-adapting detection method based on laser |
| CN107274383A (en) * | 2017-05-17 | 2017-10-20 | 南京邮电大学 | A kind of haze visibility detecting method based on deep learning |
-
2018
- 2018-07-27 CN CN201810845049.5A patent/CN109165676A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102792322A (en) * | 2009-09-30 | 2012-11-21 | 天宝导航有限公司 | Visual organization of information via associated geospatial data |
| CN103150914A (en) * | 2013-02-04 | 2013-06-12 | 安徽达尔智能控制系统有限公司 | Visual intelligent transportation meteorological monitoring and forewarning management system based on Web geographic information system (GIS) |
| CN103903008A (en) * | 2014-03-26 | 2014-07-02 | 国家电网公司 | Power transmission line fog level recognition method and system based on images |
| CN104463196A (en) * | 2014-11-11 | 2015-03-25 | 中国人民解放军理工大学 | Video-based weather phenomenon recognition method |
| CN106556579A (en) * | 2016-11-07 | 2017-04-05 | 南京理工大学 | Group's mist image all-weather self-adapting detection method based on laser |
| CN107274383A (en) * | 2017-05-17 | 2017-10-20 | 南京邮电大学 | A kind of haze visibility detecting method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 司马海峰等: "《遥感图像分类中的智能计算方法》", 31 January 2018 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886920A (en) * | 2019-01-16 | 2019-06-14 | 安徽谛听信息科技有限公司 | A kind of greasy weather stage division, greasy weather hierarchy system |
| CN109934780A (en) * | 2019-02-21 | 2019-06-25 | 北京以萨技术股份有限公司 | A kind of traffic surveillance videos defogging method based on dark primary priori |
| CN111598885A (en) * | 2020-05-21 | 2020-08-28 | 公安部交通管理科学研究所 | Automatic visibility grade marking method for highway foggy pictures |
| CN112419745A (en) * | 2020-10-20 | 2021-02-26 | 中电鸿信信息科技有限公司 | Highway group fog early warning system based on degree of depth fusion network |
| CN112749654A (en) * | 2020-12-31 | 2021-05-04 | 南京恩瑞特实业有限公司 | Deep neural network model construction method, system and device for video fog monitoring |
| CN113435405A (en) * | 2021-07-15 | 2021-09-24 | 山东交通学院 | Expressway night fog monitoring method and system based on video images |
| CN113435405B (en) * | 2021-07-15 | 2023-09-08 | 山东交通学院 | Expressway night fog monitoring method and system based on video images |
| CN115310519A (en) * | 2022-07-11 | 2022-11-08 | 广州国交润万交通信息有限公司 | A method for weather discrimination combining video and audio |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109165676A (en) | A kind of round-the-clock highway fog grade monitoring method based on video analysis | |
| CN110136447B (en) | Method for detecting lane change of driving and identifying illegal lane change | |
| Pavlic et al. | Classification of images in fog and fog-free scenes for use in vehicles | |
| CN109615862A (en) | Road vehicle movement of traffic state parameter dynamic acquisition method and device | |
| CN105760847B (en) | A kind of visible detection method of pair of helmet of motorcycle driver wear condition | |
| KR102403170B1 (en) | Apparatus for monitoring image employing to detect of vehicle number and cotrolling device | |
| CN109409337A (en) | Muck vehicle feature identification method based on convolutional neural network | |
| CN102509102A (en) | Visibility measuring method based on image study | |
| CN109409336A (en) | A kind of dense fog early warning system and method based on image recognition | |
| CN102163278A (en) | Illegal vehicle intruding detection method for bus lane | |
| CN104794906A (en) | Vehicle management platform of outdoor parking lot exit | |
| CN113888850B (en) | Sand-dust meteorological municipal road congestion situation multi-factor early warning method | |
| CN109948419A (en) | A kind of illegal parking automatic auditing method based on deep learning | |
| CN116153108A (en) | A Method for Assessing the Impact of Lighting on Intersection Safety Using a Random Forest Model | |
| CN108922172A (en) | Congestion in road based on vehicle characteristics matrix sequence mutation analysis monitors system | |
| Su et al. | A new local-main-gradient-orientation HOG and contour differences based algorithm for object classification | |
| CN112883904A (en) | Method for detecting vehicle illegal parking lane occupation | |
| CN109191492A (en) | A kind of intelligent video black smoke vehicle detection method based on edge analysis | |
| Liu et al. | Analysis of pedestrian-two-wheeler conflicts at green light digital countdown signals: A random parameter ordered logit model approach | |
| CN112883813A (en) | Expressway pavement shed object detection method based on monitoring video | |
| CN113239609A (en) | Test system and detection method for target identification and monitoring of monocular camera of intelligent vehicle | |
| Muñoz et al. | A real-time traffic alert system based on image recognition: A case of study in Spain | |
| Wibowo et al. | Implementation of background subtraction for counting vehicle using mixture of Gaussians with ROI optimization | |
| CN104715609A (en) | Passage data acquiring system of crossing with signal lamp | |
| CN115880901A (en) | Traffic violation real-time flow monitoring method and system based on bus-mounted video |
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 | ||
| CB02 | Change of applicant information |
Address after: 266500 No. 77, Lingyan Road, LINGSHANWEI sub district office, Huangdao District, Qingdao City, Shandong Province Applicant after: Issa Technology Co.,Ltd. Applicant after: QINGDAO YISA DATA TECHNOLOGY Co.,Ltd. Address before: 108, 102 / F, building 17, yard 6, Jingshun East Street, Chaoyang District, Beijing Applicant before: BEIJING YISA TECHNOLOGY Co.,Ltd. Applicant before: QINGDAO YISA DATA TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 266500 business premises, No. 77, Lingyan Road, LINGSHANWEI sub district office, Huangdao District, Qingdao City, Shandong Province Applicant after: Issa Technology Co.,Ltd. Applicant after: QINGDAO YISA DATA TECHNOLOGY Co.,Ltd. Address before: 266500 No. 77, Lingyan Road, LINGSHANWEI sub district office, Huangdao District, Qingdao City, Shandong Province Applicant before: Issa Technology Co.,Ltd. Applicant before: QINGDAO YISA DATA TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190108 |
|
| RJ01 | Rejection of invention patent application after publication |