CN104008396A - In and out people flow statistical method based on people head color and shape features - Google Patents
In and out people flow statistical method based on people head color and shape features Download PDFInfo
- Publication number
- CN104008396A CN104008396A CN201410220571.6A CN201410220571A CN104008396A CN 104008396 A CN104008396 A CN 104008396A CN 201410220571 A CN201410220571 A CN 201410220571A CN 104008396 A CN104008396 A CN 104008396A
- Authority
- CN
- China
- Prior art keywords
- target
- people
- image
- counter
- method based
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides an in and out people flow statistical method based on people head color and shape features and belongs to the technical field of video image processing. Firstly, the heads and the bodies of people are separated according to head color features, on the basis that the outlines of head areas of moving people are analyzed to be approximate circles, a people head recognition method based on Canny edge detection and based on hough transformation circle detection is proposed, and the people heads are further tracked and counted. It is shown through results that the in and out people flow statistical method based on the people head color and shape features is capable of effectively tracking and counting the people heads with high efficiency, stability and high accuracy and has high practical value.
Description
Technical field
The invention belongs to technical field of video image processing, be specifically related to a kind of import and export people flow rate statistical method based on head part's color and shape facility.
Background technology
On Vehicles Collected from Market, there are various stream of people's statistical systems to meet the different demands that use colony, mainly comprised infrared sensor scanning and counting, pressure pedal Auto-counting, the statisticals such as complicate statistics.Certain positive role has been played in the management that these several statisticals are flow of the people, but all exists multiple shortcoming separately.
Wherein, the shortcoming of infrared sensor scanning and counting is to distinguish single or many people, can not distinguish pedestrian's direct of travel; Simultaneously more responsive for environment temperature, solar radiation etc., affected greatly by external factor, counting error is large.Complicate statistics is the work of a very loaded down with trivial details and labor intensive, financial resources, in actual mechanical process, accomplish that regular, real-time, systematicness can not realize substantially.Pressure pedal Auto-counting can not be differentiated guest's turnover direction, and easily chaotic when flow of the people is large, physical unit is fragile, and installation position accuracy requirement is high, and maintainable poor, counting error is too large.
People's head inspecting method based on video analysis is the popular direction of research that the stream of people adds up field at present, and its ultimate principle is to utilize the circular feature of the number of people, head color character etc. to detect.
Based on this, the invention provides a kind of import and export people flow rate statistical method of simultaneously utilizing head part's color character and shape facility.
Summary of the invention
The technical problem to be solved in the present invention is to overcome the deficiencies in the prior art, a kind of import and export people flow rate statistical method based on head part's color and shape facility is proposed, the original video sequence image obtaining for video camera, according to number of people gray scale, present frame gray-scale map being carried out to binaryzation separates the number of people, then detect definite number of people target according to number of people boundary profile feature Canny rim detection and the circle based on hough conversion, finally follow the tracks of again, judge people into and out of behavior, realize accurate people flow rate statistical counting.
The technical solution adopted for the present invention to solve the technical problems is:
The present invention proposes a kind of import and export people flow rate statistical method based on head part's color and shape facility, and described method obtains original video sequence image using video camera as video acquisition instrument, image is analyzed, thereby completed people flow rate statistical; The described method that image is analyzed comprises the following steps:
Steps A: capturing video two field picture, and on image, choose suitable surveyed area;
Step B: convert current frame image to gray level image, then carry out binary conversion treatment, extract the number of people target in current frame image;
Step C: the image after binaryzation is carried out to region growing, each one head's target size in computed image;
Step D: the foreground target obtaining in step C is carried out to Canny rim detection and the circle detection based on hough conversion, if qualified circle detected in foreground target after, just this foreground target is carried out to picture frame processing with boundary rectangle frame, enter step e, if qualified circle do not detected in foreground target, return to steps A;
Step e: the target of carrying out picture frame processing is followed the tracks of, determine whether same target according to the coincidence area of front and back two frame target boundary rectangle frames;
Step F: the tracking results of based target is judged target turnover direction, and on video frame images, show turnover count information.
Further, the present invention is based on the import and export people flow rate statistical method of head part's color and shape facility, described video camera adopts mode fixing and that install vertically downward to obtain original video sequence image.
Further, the present invention is based on the import and export people flow rate statistical method of head part's color and shape facility, the number of people target extracting described in step B in current frame image is to obtain by number of people target area threshold values scope is set, thereby the target outside threshold values scope is foreclosed and extracts the target meeting in threshold range as number of people target.
Further, the present invention is based on the import and export people flow rate statistical method of head part's color and shape facility, in step e, be to determine whether same target by coincidence area-limit is set, if the coincidence area of the target boundary rectangle frame of front and back two frames is greater than threshold value, think same target.
Further, the present invention is based on the import and export people flow rate statistical method of head part's color and shape facility, in step F, set respectively into counter and go out counter and judge target turnover direction, if in two two field pictures of front and back target location after comparing, be into direction, enter counter and add 1, otherwise go out counter and add 1, in the time that target is left surveyed area, relatively to should target enter counter and go out the value of counter, the value that the value of counter deducts out counter if enter is for just, assert that this target is for entering direction, otherwise, be outgoing direction.
The technical solution used in the present invention compared with prior art, has following technique effect:
The present invention proposes a kind of import and export people flow rate statistical method based on head part's color and shape facility, described method can be avoided the flase drop to moving object, stationary object and the repeat count to the same number of people, and therefore the number accuracy rate of statistics is high.Conventional detection picture frame method is that the center of circle that goes out based on hough change detection and radius are drawn circle, and this method is because the number of people of motion can produce very large error, thus very unstable, cause counting precision not high.A kind of people flow rate statistical method of importing and exporting of the present invention separates the number of people according to number of people CF feature with personal, and uses rectangle frame picture frame, and more stable to the tracking of the number of people like this, it is also more accurate to count.
Brief description of the drawings
Fig. 1 is the overall flow figure that imports and exports people flow rate statistical method.
Embodiment
In order to make, of the present invention a kind of to import and export technical matters, technical scheme and technique effect that people flow rate statistical method will solve clearer, below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
Fig. 1 is the overall flow figure that imports and exports people flow rate statistical method, and a kind of people flow rate statistical method of importing and exporting, comprises target detection, target following and goal behavior analysis and the large step of stream of people's statistical counting three.
Target detection step comprises the following steps:
A. capturing video frame, reads current frame image and puts into pCapture, and pCapture is the image cursor of definition, and it is current just at processed frame that it preserves video, handles rear next frame present frame is covered, and then chooses suitable surveyed area.
B. current frame image is converted to gray level image, then carry out binary conversion treatment.It is 255 that the point that gray-scale value is less than to threshold value arranges gray-scale value, and the pixel that contrary gray-scale value is greater than threshold value is set to 0, thereby realizes binaryzation, extracts the number of people target of current frame image.
C. the image after binaryzation is carried out to region growing.Because the number of people area in surveyed area does not have too large floating, so our passing threshold limits, will foreclose with very little target very greatly.If zone bit flag, if the region area after increasing meets suitable condition, flag=1, otherwise be 0, only having is to be only satisfactory at 1 o'clock.So just obtain meeting the target of area threshold requirement.
D. the foreground target obtaining in step C is carried out to Canny rim detection and the circle detection based on hough conversion, wherein Canny rim detection is according to signal to noise ratio (S/N ratio) and location product are estimated, obtain optimization Approximation Operator, namely Canny operator, its step is to differentiate after first level and smooth finally to carry out the operation of non-maximum value inhibition; Hough conversion (Hough transformation, Hough Transform) is a kind of Feature Extraction Technology during image is processed, and it detects the object with given shape by a kind of Voting Algorithm.This process obtains a set that meets this given shape as Hough transformation result by the local maximum of calculating accumulated result in a parameter space.If qualified circle detected in foreground target after, just this foreground target is carried out to picture frame processing with boundary rectangle frame, enter step e, if qualified circle do not detected in foreground target, return to steps A.
The concrete steps of target following and goal behavior analysis and stream of people's statistical counting are as follows:
E. the target of carrying out picture frame processing is followed the tracks of.Determine whether same target according to the coincidence area of front and back two frame target boundary rectangle frames, because the displacement of front and back two frame target travels can be very not large, so the coincidence area of two target boundary rectangle frames should be larger.So, the target of front and back two frames is contrasted one by one, coincidence area is greater than thinking for same target of threshold value.
F. the tracking results of based target goes out to judge target turnover direction, and on video frame images, shows turnover count information.Set respectively into counter and go out counter and judge target turnover direction, if in two two field pictures of front and back target location after comparing, be into direction, enter counter and add 1, add 1 otherwise go out counter, in the time that target is left surveyed area, relatively to should target enter counter and go out the value of counter, the value that the value of counter deducts out counter if enter is for just, assert that this target is for entering direction, otherwise, be outgoing direction.
Obviously, it will be appreciated by those skilled in the art that a kind of disclosed people flow rate statistical method of importing and exporting of the invention described above, can also on the basis that does not depart from content of the present invention, make various improvement.Therefore, protection scope of the present invention should be determined by the content of appending claims.
Claims (5)
1. the import and export people flow rate statistical method based on head part's color and shape facility, is characterized in that: described method obtains original video sequence image using video camera as video acquisition instrument, image is analyzed, thereby completed people flow rate statistical; The described method that image is analyzed comprises the following steps:
Steps A: capturing video two field picture, and on image, choose suitable surveyed area;
Step B: convert current frame image to gray level image, then carry out binary conversion treatment, extract the number of people target in current frame image;
Step C: the image after binaryzation is carried out to region growing, each one head's target size in computed image;
Step D: the foreground target obtaining in step C is carried out to Canny rim detection and the circle detection based on hough conversion, if qualified circle detected in foreground target after, just this foreground target is carried out to picture frame processing with boundary rectangle frame, enter step e, if qualified circle do not detected in foreground target, return to steps A;
Step e: the target of carrying out picture frame processing is followed the tracks of, determine whether same target according to the coincidence area of front and back two frame target boundary rectangle frames;
Step F: the tracking results of based target is judged target turnover direction, and on video frame images, show turnover count information.
2. the import and export people flow rate statistical method based on head part's color and shape facility according to claim 1, is characterized in that: described video camera adopts mode fixing and that install vertically downward to obtain original video sequence image.
3. the import and export people flow rate statistical method based on head part's color and shape facility according to claim 1, it is characterized in that: the number of people target extracting described in step B in current frame image is to obtain by number of people target area threshold values scope is set, thereby the target outside threshold values scope is foreclosed and extracts the target meeting in threshold range as number of people target.
4. the import and export people flow rate statistical method based on head part's color and shape facility according to claim 1, it is characterized in that: in step e, be to determine whether same target by coincidence area-limit is set, if the coincidence area of the target boundary rectangle frame of front and back two frames is greater than threshold value, think same target.
5. the import and export people flow rate statistical method based on head part's color and shape facility according to claim 1, it is characterized in that: in step F, set respectively into counter and go out counter judge target pass in and out direction, if in two two field pictures of front and back target location after comparing, be into direction, enter counter and add 1, otherwise go out counter and add 1, in the time that target is left surveyed area, relatively to should target enter counter and go out the value of counter, the value that the value of counter deducts out counter if enter is for just, assert that this target is for entering direction, otherwise, be outgoing direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410220571.6A CN104008396A (en) | 2014-05-22 | 2014-05-22 | In and out people flow statistical method based on people head color and shape features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410220571.6A CN104008396A (en) | 2014-05-22 | 2014-05-22 | In and out people flow statistical method based on people head color and shape features |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104008396A true CN104008396A (en) | 2014-08-27 |
Family
ID=51369045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410220571.6A Pending CN104008396A (en) | 2014-05-22 | 2014-05-22 | In and out people flow statistical method based on people head color and shape features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104008396A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809787A (en) * | 2015-04-23 | 2015-07-29 | 中电科安（北京）系统集成有限公司 | Intelligent passenger flow volume statistics device based on camera |
| CN106778638A (en) * | 2016-12-19 | 2017-05-31 | 江苏慧眼数据科技股份有限公司 | A kind of passenger flow statistical method based on number of people detection |
| CN107316462A (en) * | 2017-08-30 | 2017-11-03 | 济南浪潮高新科技投资发展有限公司 | A kind of flow statistical method and device |
| CN111339873A (en) * | 2020-02-18 | 2020-06-26 | 南京甄视智能科技有限公司 | Passenger flow statistical method and device, storage medium and computing equipment |
| CN113762169A (en) * | 2021-09-09 | 2021-12-07 | 北京市商汤科技开发有限公司 | People flow statistical method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101847206A (en) * | 2010-04-21 | 2010-09-29 | 北京交通大学 | Pedestrian traffic statistical method and system based on traffic monitoring facilities |
| CN101872546A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Video-based method for rapidly detecting transit vehicles |
| CN103020706A (en) * | 2011-09-20 | 2013-04-03 | 佳都新太科技股份有限公司 | Visitors flow rate statistic algorithm based on moving target detection and Haar characteristics |
| CN103123726A (en) * | 2012-09-07 | 2013-05-29 | 佳都新太科技股份有限公司 | Target tracking algorithm based on movement behavior analysis |
| CN103559478A (en) * | 2013-10-07 | 2014-02-05 | 唐春晖 | Passenger flow counting and event analysis method for video monitoring of pedestrians in overlooking mode |
-
2014
- 2014-05-22 CN CN201410220571.6A patent/CN104008396A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101847206A (en) * | 2010-04-21 | 2010-09-29 | 北京交通大学 | Pedestrian traffic statistical method and system based on traffic monitoring facilities |
| CN101872546A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Video-based method for rapidly detecting transit vehicles |
| CN103020706A (en) * | 2011-09-20 | 2013-04-03 | 佳都新太科技股份有限公司 | Visitors flow rate statistic algorithm based on moving target detection and Haar characteristics |
| CN103123726A (en) * | 2012-09-07 | 2013-05-29 | 佳都新太科技股份有限公司 | Target tracking algorithm based on movement behavior analysis |
| CN103559478A (en) * | 2013-10-07 | 2014-02-05 | 唐春晖 | Passenger flow counting and event analysis method for video monitoring of pedestrians in overlooking mode |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809787A (en) * | 2015-04-23 | 2015-07-29 | 中电科安（北京）系统集成有限公司 | Intelligent passenger flow volume statistics device based on camera |
| CN106778638A (en) * | 2016-12-19 | 2017-05-31 | 江苏慧眼数据科技股份有限公司 | A kind of passenger flow statistical method based on number of people detection |
| CN107316462A (en) * | 2017-08-30 | 2017-11-03 | 济南浪潮高新科技投资发展有限公司 | A kind of flow statistical method and device |
| CN111339873A (en) * | 2020-02-18 | 2020-06-26 | 南京甄视智能科技有限公司 | Passenger flow statistical method and device, storage medium and computing equipment |
| CN111339873B (en) * | 2020-02-18 | 2021-04-20 | 南京甄视智能科技有限公司 | Passenger flow statistical method and device, storage medium and computing equipment |
| CN113762169A (en) * | 2021-09-09 | 2021-12-07 | 北京市商汤科技开发有限公司 | People flow statistical method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10445567B2 (en) | Pedestrian head identification method and system | |
| CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| CN106128121B (en) | Vehicle queue length fast algorithm of detecting based on Local Features Analysis | |
| US8170278B2 (en) | System and method for detecting and tracking an object of interest in spatio-temporal space | |
| CN104318263A (en) | Real-time high-precision people stream counting method | |
| CN103150549B (en) | A kind of road tunnel fire detection method based on the early stage motion feature of smog | |
| CN101996310B (en) | Face detection and tracking method based on embedded system | |
| CN105069408A (en) | Video portrait tracking method based on human face identification in complex scenario | |
| CN104933710A (en) | Intelligent analysis method of store people stream track on the basis of surveillance video | |
| CN104008396A (en) | In and out people flow statistical method based on people head color and shape features | |
| CN106022231A (en) | Multi-feature-fusion-based technical method for rapid detection of pedestrian | |
| CN102214291A (en) | Method for quickly and accurately detecting and tracking human face based on video sequence | |
| CN103310194A (en) | Method for detecting head and shoulders of pedestrian in video based on overhead pixel gradient direction | |
| Li et al. | A people counting method based on head detection and tracking | |
| CN103218605A (en) | Quick eye locating method based on integral projection and edge detection | |
| CN107516423B (en) | Video-based vehicle driving direction detection method | |
| CN104599291B (en) | Infrared motion target detection method based on structural similarity and significance analysis | |
| CN102693427A (en) | Method and device for forming detector for detecting images | |
| CN105426816A (en) | Method and device of processing face images | |
| CN104123714A (en) | Optimal target detection scale generation method in people flow statistics | |
| CN103049788A (en) | Computer-vision-based system and method for detecting number of pedestrians waiting to cross crosswalk | |
| Li et al. | Automatic passenger counting system for bus based on RGB-D video | |
| WO2021022698A1 (en) | Following detection method and apparatus, and electronic device and storage medium | |
| CN104899559A (en) | Rapid pedestrian detection method based on video monitoring | |
| CN104268902A (en) | Multi-target video tracking method for industrial site |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140827 |