CN105959639A - Pedestrian monitoring method in urban street area based on ground calibration - Google Patents
Pedestrian monitoring method in urban street area based on ground calibration Download PDFInfo
- Publication number
- CN105959639A CN105959639A CN201610395781.8A CN201610395781A CN105959639A CN 105959639 A CN105959639 A CN 105959639A CN 201610395781 A CN201610395781 A CN 201610395781A CN 105959639 A CN105959639 A CN 105959639A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- region
- monitoring
- avenue
- activity
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
Abstract
The present invention provides a pedestrian monitoring method in an urban street area based on ground calibration. The method comprises the steps of (1) dividing a target area, wherein the ground area of pedestrian movement is divided from the monitoring scene in the urban street area as a target area to carry out ground calibration, and (2) monitoring the target area, wherein a camera for shooting the target area is connected to a background server, the server is provided with a positioning device, a recording device, an analysis device and a control device, the positioning device is used for positioning the target area, the recording device is used for recording the pedestrian movement comprising shapes, colors, movement states and attitude information in the target area, and the analysis device is used for analyzing the characteristics of pedestrians, and the characteristics comprise shapes, colors, movement states and attitude information.
Description
Technical field
The present invention relates to the use of the technology that pedestrian in avenue region is monitored by photographic head,
In a kind of avenue region based on ground demarcation, pedestrian monitors method.
Background technology
Along with the development of information technology and universal, photographic head in avenue region in obtain
Large-scale laying.Utilize the monitoring that these photographic head are carried out, to daily in avenue region
The guarantee of the safety of life serves the most important effect.In concrete monitoring, to city
In region, city street, the monitoring of pedestrian is the most of paramount importance one piece of content.But it is because city
In region, city street, the graphical analysis of the monitoring of pedestrian is more complicated, is difficult to accomplish to process in real time.
The at present utilization to camera head monitor has thing there occurs to consult use often, and do so has
Time efficiency and effect be not it is obvious that do not reach and process in time and the effect of accident prevention.
Improve operating rate and the efficiency of pedestrian's monitoring in avenue region at present so that it is have more reality
Time property is the development trend monitoring work in avenue region.
As a example by Fig. 1, we can find in whole avenue region in monitored picture, institute
The scope of monitoring is very big, but pedestrian institute can the region (such as black line institute target area in figure) of activity
Only account for very small part therein.If to whole when of monitoring in avenue region
Individual region is analyzed to waste substantial amounts of resource and time with target pedestrian's searching, makes pedestrian supervise
Ore-controlling Role loses the effect of real-time.If but we monitor in avenue region and regard
The when of the analysis that frequency is carried out, the scope being only likely to occur monitored picture pedestrian is counted
Calculate, so can improve work efficiency to a great extent, improve the real-time of system.
In view of discussed above, it is proposed that row in a kind of avenue region based on ground demarcation
People monitors technical scheme.
Summary of the invention
The present invention is to solve drawbacks described above and deficiency present in prior art, it is provided that a kind of
Monitor method based on pedestrian in the avenue region that ground is demarcated, the method monitoring efficiency high and
Result precision is high, and effectively reduces the waste to resource, and real-time is high.
For solving above-mentioned technical problem, the present invention provides a kind of avenue demarcated based on ground
In region, pedestrian monitors method, comprises the following steps:
Step one, segmentation object region: be partitioned in the monitoring scene in avenue region
The ground areas of pedestrian activity, carries out ground demarcation as target area;
Step 2, is monitored target area: by for photographic subjects region photographic head with
Background server is connected, and server is provided with positioner, recording equipment, analytical equipment and control
Device, positioner is used for positioning target area;Recording equipment is in record object region
Pedestrian activity's situation, including shape color, kinestate and azimuth information;Analytical equipment is used
In the feature of analysis pedestrian, including shape, color, kinestate and azimuth information.
Wherein, in described step one, during segmentation object region, use Algorithms for Automatic Classification to prison
The ground areas of the pedestrian activity in control scene is demarcated, and target area need to meet two bars
Part: one, target area is in the effective monitoring region of photographic head;Two, institute is contained in target area
There is the range of activity region of pedestrian.
The division methods of pedestrian activity range areas comprises the following steps:
A. the shade of the monitoring scene video in avenue region is eliminated, formed original
The background interface of video area;
B. to the background interface of the monitoring scene video in avenue region with pedestrian activity
Video clip be analyzed, determine the activity trail of pedestrian;
C. the work of pedestrian in the monitoring scene in avenue region is determined by the activity trail of pedestrian
Dynamic scope, and carry out ground demarcation, the region of demarcation is the target area of pedestrian activity.
Step a is specifically, the monitoring background video first obtained in one section of avenue region is made
For original video, including n frame background picture, wherein, n is the monitoring background video obtained
The number of middle frame;Each background picture is replaced the most again by the meansigma methods of n frame background picture pixel
Each color vector of middle pixel, forms new pixel, reaches to eliminate the purpose of shade,
Thus form original background interface.
Step b is specifically, with the set of frame of pedestrian activity in setting video
S:{s1,s2,s3...sm, then the set A of the activity trail of pedestrian, computing formula is as follows:
A:{a1,a2,a3...am}=b-S:{s1,s2,s3...sm}
Wherein, b is the background interface of video.
Specifically, first the color space of the video clip with pedestrian's vestige, is turned by step c
Being changed to HSL space, conversion formula is as follows:
Wherein, in HSL space, three component: h are form and aspect, and s is saturation, and l is brightness;
In colored rgb space, three component: r represent redness;G represents green;B represents blueness;
Then, obtain meansigma methods H of pedestrian vestige HSL space three components of pixel, S and L,
And the pixel at background interface and this meansigma methods are carried out computing acquisition ground demarcation, operational formula is such as
Under:
abs(H-h)<1 and abs(S-s)<1 and abs(L-l)<10
Operation result is the target area of pedestrian activity's vestige.
The Advantageous Effects that the present invention is reached: the present invention provides a kind of based on ground demarcation
In avenue region, pedestrian monitors method, and the method monitoring efficiency is high and result precision is high,
And effectively reducing the waste to resource, real-time is high.
Accompanying drawing explanation
Fig. 1 schematic flow sheet of the present invention;
Fig. 2 embodiments of the present invention original background frame;
Fig. 3 embodiments of the present invention pedestrian activity's frame;
Fig. 4 embodiments of the present invention carry out shadow removing to original background frame and does not carries out the moon
Shadow eliminates comparison diagram;
The pedestrian activity's vestige figure extracted in Fig. 5 embodiments of the present invention.
Detailed description of the invention
The invention will be further described below in conjunction with the accompanying drawings.Following example are only used for more clear
Chu's ground explanation technical scheme, and can not limit the scope of the invention with this.
As it is shown in figure 1, the present invention provides row in a kind of avenue region based on ground demarcation
People monitors method, comprises the following steps:
Step one, segmentation object region: be partitioned in the monitoring scene in avenue region
The ground areas of pedestrian activity, carries out ground demarcation as target area;
During segmentation object region, use Algorithms for Automatic Classification to the pedestrian activity's in monitoring scene
Ground areas is demarcated, and target area need to meet two conditions: one, and target area is in shooting
In the effective monitoring region of head;Two, the range of activity region of all pedestrians is contained in target area.
The division methods of pedestrian activity range areas comprises the following steps:
A. the shade of the monitoring scene video in avenue region is eliminated, formed original
The background interface of video area, shade not only has influence on ground and demarcates, also can be to the monitoring of pedestrian
Analysis brings difficulty;As embodiments of the invention, as in figure 2 it is shown, be the monitoring of a certain street
The original background frame of video.
First the monitoring background video in one section of avenue region is obtained as original video, bag
Containing n frame background picture;Each back of the body is replaced the most again by the meansigma methods of n frame background picture pixel
In scape picture, each color vector of pixel, forms new pixel, reaches to eliminate shade
Purpose, thus form original background interface.As shown in Figure 4, wherein, M is eradicating efficacy
Not eliminating the background interface of shade, N is the background interface after eliminating shade.
B. to the background interface of the monitoring scene video in avenue region with pedestrian activity
Video clip be analyzed, determine the activity trail of pedestrian, as shown in Figure 3;
With the set S:{s of frame of pedestrian activity in setting video1,s2,s3...sm, then the work of pedestrian
The set A of dynamic vestige, computing formula is as follows:
A:{a1,a2,a3...am}=b-S:{s1,s2,s3...sm}
Wherein, b is the background interface of video.
C. the work of pedestrian in the monitoring scene in avenue region is determined by the activity trail of pedestrian
Dynamic scope, and carry out ground demarcation, the region of demarcation is the target area of pedestrian activity, such as figure
Shown in 5;
First, it is HSL space by the color space conversion of the video clip with pedestrian's vestige,
Conversion formula is as follows:
Wherein, in HSL space, three component: h are form and aspect, and s is saturation, and l is brightness;
In colored rgb space, three component: r represent redness;G represents green;B represents blueness;
Then, obtain meansigma methods H of pedestrian vestige HSL space three components of pixel, S and L,
And the pixel at background interface and this meansigma methods are carried out computing acquisition ground demarcation, operational formula is such as
Under:
abs(H-h)<1 and abs(S-s)<1 and abs(L-l)<10
Operation result is the target area of pedestrian activity's vestige.
Step 2, is monitored target area: by for photographic subjects region photographic head with
Background server is connected, and server is provided with positioner, recording equipment, analytical equipment and control
Device, positioner is used for positioning target area;Recording equipment is in record object region
Pedestrian activity's situation, including shape color, kinestate and azimuth information;Analytical equipment is used
In the feature of analysis pedestrian, including shape, color, kinestate and azimuth information.
In order to better illustrate the Advantageous Effects of the present invention, in conjunction with the embodiments such as Fig. 2-5 institute
The scene shown is described in more detail:
Fig. 2 is the original background frame of a certain street monitor video, and the picture of Fig. 2 is carried out shade
Eliminate and do not carry out the contrast of shadow removing as shown in Figure 4, by extracting after carrying out shadow removing
The pedestrian activity's vestige with pedestrian activity shown in Fig. 3, as shown in Figure 5.Result shows,
The method monitoring efficiency is high and result precision is high, picture is clear.
The above is only the preferred embodiment of the present invention, it is noted that lead for this technology
For the those of ordinary skill in territory, on the premise of without departing from the technology of the present invention principle, it is also possible to
Making some improvement and deformation, these improve and deformation also should be regarded as protection scope of the present invention.
Claims (6)
1. in avenue region based on ground demarcation, pedestrian monitors method, it is characterised in that
Comprise the following steps:
Step one, segmentation object region: be partitioned in the monitoring scene in avenue region
The ground areas of pedestrian activity, carries out ground demarcation as target area;
Step 2, is monitored target area: by for photographic subjects region photographic head with
Background server is connected, and server is provided with positioner, recording equipment, analytical equipment and control
Device, positioner is used for positioning target area;Recording equipment is in record object region
Pedestrian activity's situation, including shape color, kinestate and azimuth information;Analytical equipment is used
In the feature of analysis pedestrian, including shape, color, kinestate and azimuth information.
Pedestrian's monitoring in avenue region based on ground demarcation the most according to claim 1
Method, it is characterised in that: in described step one, during segmentation object region, use classification automatically
The ground areas of the pedestrian activity in monitoring scene is demarcated by algorithm, and target area need to meet
Two conditions: one, target area is in the effective monitoring region of photographic head;Two, target area
Contain the range of activity region of all pedestrians.
Pedestrian's monitoring in avenue region based on ground demarcation the most according to claim 2
Method, it is characterised in that: the division methods of pedestrian activity range areas comprises the following steps:
A. the shade of the monitoring scene video in avenue region is eliminated, formed original
The background interface of video area;
B. to the background interface of the monitoring scene video in avenue region with pedestrian activity
Video clip be analyzed, determine the activity trail of pedestrian;
C. the work of pedestrian in the monitoring scene in avenue region is determined by the activity trail of pedestrian
Dynamic scope, and carry out ground demarcation, the region of demarcation is the target area of pedestrian activity.
Pedestrian's monitoring in avenue region based on ground demarcation the most according to claim 3
Method, it is characterised in that: first step a specifically, obtain in one section of avenue region
Monitoring background video, as original video, includes n frame background picture, and wherein, n is for obtaining
Monitoring background video in the number of frame;The most again with the meansigma methods generation of n frame background picture pixel
For each color vector of pixel in each background picture, form new pixel, reach to disappear
Except the purpose of shade, thus form original background interface.
Pedestrian's monitoring in avenue region based on ground demarcation the most according to claim 3
Method, it is characterised in that: step b is specifically, with the collection of frame of pedestrian activity in setting video
Close S:{s1,s2,s3...sm, then the set A of the activity trail of pedestrian, computing formula is as follows:
A:{a1,a2,a3...am}=b-S:{s1,s2,s3...sm}
Wherein, b is the background interface of video.
Pedestrian's monitoring in avenue region based on ground demarcation the most according to claim 3
Method, it is characterised in that: step c specifically, first, by video circle with pedestrian's vestige
The colored rgb space in face is converted to HSL space, and conversion formula is as follows:
Wherein, in HSL space, three component: h are form and aspect, and s is saturation, and l is brightness;
In colored rgb space, three component: r represent redness;G represents green;B represents blueness;
Then, obtain meansigma methods H of pedestrian vestige HSL space three components of pixel, S and L,
And the pixel at background interface and this meansigma methods are carried out computing acquisition ground demarcation, operational formula is such as
Under:
abs(H-h)<1 and abs(S-s)<1 and abs(L-l)<10
Operation result is the target area of pedestrian activity's vestige.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610395781.8A CN105959639B (en) | 2016-06-06 | 2016-06-06 | Pedestrian's monitoring method in avenue region based on ground calibration |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610395781.8A CN105959639B (en) | 2016-06-06 | 2016-06-06 | Pedestrian's monitoring method in avenue region based on ground calibration |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105959639A true CN105959639A (en) | 2016-09-21 |
CN105959639B CN105959639B (en) | 2019-06-14 |
Family
ID=56908781
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610395781.8A Active CN105959639B (en) | 2016-06-06 | 2016-06-06 | Pedestrian's monitoring method in avenue region based on ground calibration |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105959639B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108491818A (en) * | 2018-03-30 | 2018-09-04 | 北京三快在线科技有限公司 | Detection method, device and the electronic equipment of target object |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102201059A (en) * | 2011-05-20 | 2011-09-28 | 北京大学深圳研究生院 | Pedestrian detection method and device |
CN102332157A (en) * | 2011-06-15 | 2012-01-25 | 夏东 | Method for eliminating shadow |
JP2012242993A (en) * | 2011-05-18 | 2012-12-10 | Nissan Motor Co Ltd | Moving body monitoring device and moving body monitoring method |
CN102930719A (en) * | 2012-10-09 | 2013-02-13 | 北京航空航天大学 | Video image foreground detection method for traffic intersection scene and based on network physical system |
CN103324913A (en) * | 2013-05-29 | 2013-09-25 | 长安大学 | Pedestrian event detection method based on shape features and trajectory analysis |
CN103400371A (en) * | 2013-07-09 | 2013-11-20 | 河海大学 | Multi-camera synergistic monitoring equipment and method |
-
2016
- 2016-06-06 CN CN201610395781.8A patent/CN105959639B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012242993A (en) * | 2011-05-18 | 2012-12-10 | Nissan Motor Co Ltd | Moving body monitoring device and moving body monitoring method |
CN102201059A (en) * | 2011-05-20 | 2011-09-28 | 北京大学深圳研究生院 | Pedestrian detection method and device |
CN102332157A (en) * | 2011-06-15 | 2012-01-25 | 夏东 | Method for eliminating shadow |
CN102930719A (en) * | 2012-10-09 | 2013-02-13 | 北京航空航天大学 | Video image foreground detection method for traffic intersection scene and based on network physical system |
CN103324913A (en) * | 2013-05-29 | 2013-09-25 | 长安大学 | Pedestrian event detection method based on shape features and trajectory analysis |
CN103400371A (en) * | 2013-07-09 | 2013-11-20 | 河海大学 | Multi-camera synergistic monitoring equipment and method |
Non-Patent Citations (1)
Title |
---|
朱琳: "场景信息估计与遮挡条件下目标跟踪问题研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108491818A (en) * | 2018-03-30 | 2018-09-04 | 北京三快在线科技有限公司 | Detection method, device and the electronic equipment of target object |
CN108491818B (en) * | 2018-03-30 | 2019-07-05 | 北京三快在线科技有限公司 | Detection method, device and the electronic equipment of target object |
Also Published As
Publication number | Publication date |
---|---|
CN105959639B (en) | 2019-06-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021208275A1 (en) | Traffic video background modelling method and system | |
CN109657575B (en) | Intelligent video tracking algorithm for outdoor constructors | |
CN104050480A (en) | Cigarette smoke detection method based on computer vision | |
CN105469105A (en) | Cigarette smoke detection method based on video monitoring | |
CN107566781A (en) | Video frequency monitoring method and video monitoring equipment | |
CN112149761B (en) | Electric power intelligent construction site violation detection method based on YOLOv4 improved algorithm | |
CN111414887A (en) | Secondary detection mask face recognition method based on YO L OV3 algorithm | |
CN109918971B (en) | Method and device for detecting number of people in monitoring video | |
CN103985130B (en) | A kind of saliency analysis method for complex texture image | |
CN103810707B (en) | A kind of image vision significance detection method based on moving-vision focus | |
CN108039044A (en) | The system and method that Vehicular intelligent based on multiple dimensioned convolutional neural networks is lined up | |
CN116152748B (en) | River and lake supervision method and system based on blue algae identification | |
CN106228157A (en) | Coloured image word paragraph segmentation based on image recognition technology and recognition methods | |
CN113887412A (en) | Detection method, detection terminal, monitoring system and storage medium for pollution emission | |
CN106682596A (en) | Violation fishing monitoring method based on video image analysis | |
CN111460988A (en) | Illegal behavior identification method and device | |
CN107491764A (en) | A kind of violation based on depth convolutional neural networks drives detection method | |
CN112818953A (en) | Mask wearing state identification method, device, equipment and readable storage medium | |
CN104318240A (en) | Flower bud discriminating method based on computer vision | |
CN106780540A (en) | Towards the ground cloud atlas cloud layer tracking of photovoltaic generation and method for early warning | |
CN103295238B (en) | Video real-time location method based on ROI motion detection on Android platform | |
CN106023199A (en) | Image analysis technology-based fume blackness intelligent detection method | |
CN106372566A (en) | Digital signage-based emergency evacuation system and method | |
CN105959639A (en) | Pedestrian monitoring method in urban street area based on ground calibration | |
CN104182976A (en) | Field moving object fining extraction 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 | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20211229 Address after: 214191 unit 3, jinhuwan Shijia, Dongbei Tang, Xishan District, Wuxi City, Jiangsu Province Patentee after: WUXI XIELIAN INFORMATION TECHNOLOGY CO.,LTD. Address before: 1 No. 211167 Jiangsu city of Nanjing province Jiangning Science Park Hongjing Road Patentee before: NANJING INSTITUTE OF TECHNOLOGY |