CN102750539B - A kind of device that regional space crowd is added up and statistical method thereof - Google Patents

A kind of device that regional space crowd is added up and statistical method thereof Download PDF

Info

Publication number
CN102750539B
CN102750539B CN201210184372.5A CN201210184372A CN102750539B CN 102750539 B CN102750539 B CN 102750539B CN 201210184372 A CN201210184372 A CN 201210184372A CN 102750539 B CN102750539 B CN 102750539B
Authority
CN
China
Prior art keywords
human body
module
crowd
step
target
Prior art date
Application number
CN201210184372.5A
Other languages
Chinese (zh)
Other versions
CN102750539A (en
Inventor
何玉成
Original Assignee
深圳市海亿达能源科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市海亿达能源科技股份有限公司 filed Critical 深圳市海亿达能源科技股份有限公司
Priority to CN201210184372.5A priority Critical patent/CN102750539B/en
Publication of CN102750539A publication Critical patent/CN102750539A/en
Application granted granted Critical
Publication of CN102750539B publication Critical patent/CN102750539B/en

Links

Abstract

A kind of device that regional space crowd is added up comprises the acquisition module (10), video processing module (20), rejecting module (30), memory module (40), human body target identification module (50), data processing module (60) and the data outputting module (70) that connect successively. First gather image information and the infrared temperature information of surveillance area simultaneously; According to infrared temperature information, reject the target that does not meet human body temperature in guarded region; Human body in guarded region is identified; Identified human body quantity is carried out to statistical computation, and result is exported. It is more simple and easy to use that this device is compared system, and using method is easily identified by computer system, and detection speed is fast, and committed memory is few, can be effectively, the crowd's quantity in statistical space region fast and accurately.

Description

A kind of device that regional space crowd is added up and statistical method thereof

Technical field

The present invention relates to data processing technique, particularly crowd's quantity statistics device and statistical method in area of space.

Background technology

The today of day by day improving in information system management level, carry out crowd's quantity statistics for huge places of flow of the people such as station, market, supermarkets, and according to the volume of the flow of passengers number, can reasonably reduce the large energy consumption equipment such as opening amount of air-conditioning, be that economy is energy-conservation again like this, but also environmental protection. For this problem, the detection of crowd's quantity has been played key effect in automatic video monitoring, interpersonal many-side such as mutual in applying. Traditional crowd's statistical method realizes based on artificial technology or artificial electronic equipment flip-flop number. The method not only workload is large, and wrong report and rate of failing to report are high, and efficiency is very low, and is subject to equipment and staff's restriction, and range of application is extremely limited. Along with the development of computer technology, utilize computer vision and image processing techniques, realize video crowd demographics, in actual public administration field, be widely used. At present, the crowd's statistical method based on video monitoring has mainly been followed the general step of crowd's detection, tracking technique technically. The method is single in background, in the lower situation of crowd density, especially for single alone by a certain region, there is good statistics effect. But, because China is populous nation, crowd's One's name is legion, and the feature such as mobility is larger. Therefore, for research, a kind of to carry out statistic device for regional space crowd imperative.

Summary of the invention

The invention provides a kind of device that regional space crowd is added up and statistical method thereof, adopt multiple RM combination and automatically add up, solve the drawback that can not add up large crowd in prior art.

The present invention comprises for this device that regional space crowd is added up providing is provided the acquisition module, video processing module, rejecting module, memory module, human body target identification module, data processing module and the data outputting module that connect successively. Described acquisition module comprises the video acquisition module and the infrared temperature detecting module that connect respectively video processing module; Described human body target identification module comprises image comparison module, Head recognition module and human-shaped body identification module; This device also comprises data disaply moudle and the data communication module of connection data output module respectively.

And the statistical method of this device comprises the following steps: A. gathers image information and the infrared temperature information of surveillance area first simultaneously; B. color image information is converted to gray level image information; C. according to infrared temperature information in the guarded region gathering in steps A, reject the target that does not meet human body temperature in guarded region; D. the human body in guarded region is identified; E. identified human body quantity is carried out to statistical computation, and result is exported.

The collection period of image information described in steps A and infrared temperature information can arrange any a period of time taking millisecond, second, minute, hour as unit, human body identification in step D comprises step by step following: D1. worked as for two image frame grabber times very in short-term, during as 1ms, 10ms, can adopt front and back two two field pictures in same area are contrasted, reject repeater's body target; D2. to identifying according to number of people feature in surveillance area, and the mark that carries out of human body will be defined as; D3. the part not being labeled is carried out to characteristics of human body's identification.

The identification of human body in step D2 and D3 is taking the center in surveillance area as the center of circle, with concentric circles form around the outside scanning search of this center of circle from coil to coil. In step e, the computational methods of human body quantity are that fixed human body in step D is accumulated. The output of data described in step e comprises by display screen and shows statistics, statistics is outputed in other equipment or background management system by data-out port.

The present invention mainly by camera acquisition to image process, analyze and rejecting etc. that thermal camera carries out pseudo-human body target finally realizes the crowd in area-of-interest is added up, can be for crowd's quantity real-time statistics, therefore, can be used as the strong foundation of crowd's flow-control. Utilize a kind of device that regional space crowd is added up of the present invention and statistical method thereof to play critical effect in the place that the crowd is dense such as waiting room, passenger station at the train station. The advantages such as it is simpler that this device is compared system, easy to use, and using method is easily identified by computer system, and detection speed is fast, and committed memory is few. Can be effective, fast, add up accurately crowd's quantity of railway station waiting room, also can be according to crowd's quantity of calculating, the equipment of reasonably closing the large energy consumptions such as some unwanted air-conditionings, has not only ensured personal safety, has also reached the object of energy-saving and emission-reduction, when having met the trend of national development, also reach the effect of environmental protection. Its utilization has following beneficial effect:

1, this invention is a device, compares system simpler, easy to use.

2, this device adopts image-capture mode, the crawl time can arrange, as every millisecond, per second, per minute etc., can be by the candid photograph time of setting, grasp the situation of change of the crowd's quantity in a period of time, reach real-time Monitoring Population quantity, and then reasonably control the volume of the flow of passengers, ensure personal safety.

3, the method has improved real-time, and noise resisting ability is strong and committed memory is little, precision advantages of higher.

4, for highdensity place on a large scale, still accomplish high precision crowd quantity statistics, play immeasurable value for controlling crowd's quantity, not only ensured personal safety, can also be as the effective reference frame of large energy consumption equipment use amount, i.e. energy-conservation environmental protection again.

Brief description of the drawings

Fig. 1 is the schematic flow sheet of statistical method of the present invention. Fig. 2 is the decomposition step schematic diagram of step D of the present invention. Fig. 3 is a kind of schematic diagram that regional space crowd is carried out to statistic device of the present invention.

Detailed description of the invention

In conjunction with above-mentioned brief description of the drawings specific embodiments of the invention.

In Fig. 3, a kind of device that regional space crowd is added up of the present invention, is characterized in that: this device comprises the acquisition module 10, video processing module 20, rejecting module 30, memory module 40, human body target identification module 50, data processing module 60 and the data outputting module 70 that connect successively. Described acquisition module 10 comprises the video acquisition module 11 and the infrared temperature detecting module 12 that connect respectively video processing module 20; Described human body target identification module 50 comprises image comparison module 51, Head recognition module 52 and human-shaped body identification module 53; This device also comprises data disaply moudle 80 and the data communication module 90 of connection data output module 70 respectively. Wherein: video acquisition module 11 moves crowd's video image of monitoring in real time for obtaining from image capture device video camera to regional space; Infrared temperature detecting module 12 gathers for the heat radiation situation to view data; Video processing module 20 is for being converted into greyscale video image by colored guarded region image; Reject module 30 for distinguishing human body target and non-human target, and do not meet the pseudo-human body target of body temperature by the heat radiation situation rejecting heat in the view data of thermal camera; Memory module 40 is for storing the Human form models of rejecting the greyscale video image after non-human target and can represent majority's all ages and classes section; Human body target identification module 50(comprises image comparison module 51, Head recognition module 52 and human-shaped body identification module 53):

Image comparison module 51 contrasts identification to two adjacent two field pictures, has ensured the accuracy of Output rusults, has effectively rejected repeater's body target. The application of the method was only suitable for being applied in two frame period times less time, and for example two frame times are spaced apart in the situation in 1ms, 10ms, otherwise, the judgement that can make the mistake to Output rusults by the method; The size at this device two frame time intervals that identification arranges automatically, if the time interval is larger, internal algorithm will not adopt this method.

Head recognition module 52 is rejected non-human target and is repeated to identify more accurately under the condition of human body target of human body target the quantity of the number of people above, saved the time of human body search, does accurate preparation for counting the number of people;

Human-shaped body identification module 53 (becomes the black picture element of Head recognition human body target out into white pixel for the human body target whiteness to the easy identification having come out, prevent from repeating to add up human body target) after image, carry out the identification of human figure, finally identify this regional space all people body target; Data processing module 60 is for carrying out number accumulation algorithm by the image of finally handling well; Data outputting module 70, for by this range statistics crowd's quantity out, is exported; Data disaply moudle 80 is for showing crowd's statistics of result output module; Data communication module 90 is connected in background monitoring system, realizes data upload.

In Fig. 1 and Fig. 2, the statistical method of said apparatus comprises the following steps: A. gathers image information and the infrared temperature information of surveillance area first simultaneously; From image capture device video camera, obtain selecting to need the video area of monitoring in the video image of mobile crowd monitoring in real time, and obtain the data of this region infrared emanation situation; B. color image information is converted to gray level image information; C. according to infrared temperature information in the guarded region gathering in steps A, reject the target that does not meet human body temperature in guarded region; D. the human body in guarded region is identified; E. identified human body quantity is carried out to statistical computation, and result is exported.

The collection period of image information described in steps A and infrared temperature information can be set to taking millisecond, second, minute for example, as cycle time (10ms, 30s, 1min etc.) of unit, the identification of human body in step D comprises step by step following: D1. when signal acquisition periods be between 1ms to 10ms time, front and back two two field pictures in same area are contrasted, reject repeater's body target, when every two two field picture interval times are when long, can affect on the contrary the precision of this region crowd statistics, therefore, will walk around in this case this step.

The identification of human body in step D also comprises step by step following: D2. is to identifying according to number of people feature in surveillance area, and will be defined as the mark that carries out of human body; In the new region of human body target, there are some black picture elements to assemble shape matching rule, be equivalent to similar round, adopt Head recognition method just can determine easily human body target; D3. the part not being labeled is carried out to characteristics of human body's identification, the black picture element whiteness of definite human body target, prevent from repeating to add up crowd's quantity, the human figure image that can represent majority of the black pixel point in remaining human body target region and storage is analyzed, finally determines human body target.

The identification of human body in step D2 and D3 is that center using surveillance area is as the center of circle, with concentric circles form around the outside scanning search of this center of circle from coil to coil, at step C according to human body temperature, reject the region that does not have human body target in region, obtain a new human body target region having dwindled, can greatly dwindle like this hunting zone of human body target, and then save search time, improve the efficiency of crowd's statistics, area image is cut apart, taking any point (preferably center) of new region as the center of circle, evenly expand round area to circumferential outer edges gradually with concentric circles successively and carry out human body target search, area image is effectively cut apart, as starting point, be assumed to be vectorial starting point from the center of new region, according to certain radius length, the length of vector, as Second hand for clock and watch, carries out circular scanning counterclockwise, the length that evenly expands gradually vectorial radius, search area expands gradually. by above step, human body target identification is carried out in the new human body target region forming, finally count this regional space crowd's total number of persons.

In step e, the computational methods of human body quantity are that fixed human body in step D is accumulated, and the output of data described in step e comprises by display screen and shows statistics, statistics is outputed in other equipment or background management system by data-out port.

Above content is in conjunction with concrete preferred embodiment further description made for the present invention, can not assert that specific embodiment of the invention is confined to these explanations. For general technical staff of the technical field of the invention, without departing from the inventive concept of the premise, can also make some simple deduction or replace, all should be considered as belonging to protection scope of the present invention.

Claims (7)

1. the statistical method that regional space crowd is carried out, the method comprises the following steps: A. gathers image information and the infrared temperature information of surveillance area first simultaneously, and the collection period of described image information and infrared temperature information can arrange any a period of time taking millisecond, second, minute, hour as unit; B. color image information is converted to gray level image information; C. according to infrared temperature information in the guarded region gathering in steps A, reject the target that does not meet human body temperature in guarded region; D. the human body in guarded region is identified, it is step by step following that human body identification comprises: D1., when signal acquisition periods is between 1ms to 10ms time, contrasts rejecting repeater body target to two two field pictures before and after in same area; D2. to identifying according to number of people feature in surveillance area, and the human body target black picture element whiteness of human body will be defined as; D3. the part not being labeled is carried out to characteristics of human body's identification, the human figure image that can represent majority of the black pixel point in remaining human body target region and storage is analyzed, finally determine human body target; E. identified human body quantity is carried out to statistical computation, and result is exported.
2. statistical method according to claim 1, is characterized in that: the identification of human body in step D2 and D3 be geometric center taking surveillance area as the center of circle, with concentric circles form around the outside scanning search of this center from coil to coil.
3. statistical method according to claim 1, is characterized in that: in step e, the computational methods of human body quantity are that fixed human body in step D is accumulated.
4. statistical method according to claim 1, is characterized in that: the output described in step e comprises by display screen and shows statistics, statistics is outputed in other equipment or background management system by data-out port.
5. one kind according to described in any one in claim 1 to 4, regional space crowd being carried out to the constructed device of statistical method, it is characterized in that: this device comprises the acquisition module (10), video processing module (20), rejecting module (30), memory module (40), human body target identification module (50), data processing module (60) and the data outputting module (70) that connect successively, and described human body target identification module (50) comprises image comparison module (51), Head recognition module (52) and human-shaped body identification module (53).
6. device according to claim 5, is characterized in that: described acquisition module (10) comprises the video acquisition module (11) and the infrared temperature detecting module (12) that connect respectively video processing module (20).
7. device according to claim 5, is characterized in that: this device also comprises data disaply moudle (80) and the data communication module (90) of connection data output module (70) respectively.
CN201210184372.5A 2012-06-06 2012-06-06 A kind of device that regional space crowd is added up and statistical method thereof CN102750539B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210184372.5A CN102750539B (en) 2012-06-06 2012-06-06 A kind of device that regional space crowd is added up and statistical method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210184372.5A CN102750539B (en) 2012-06-06 2012-06-06 A kind of device that regional space crowd is added up and statistical method thereof

Publications (2)

Publication Number Publication Date
CN102750539A CN102750539A (en) 2012-10-24
CN102750539B true CN102750539B (en) 2016-05-18

Family

ID=47030704

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210184372.5A CN102750539B (en) 2012-06-06 2012-06-06 A kind of device that regional space crowd is added up and statistical method thereof

Country Status (1)

Country Link
CN (1) CN102750539B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103297755A (en) * 2013-05-13 2013-09-11 广东欧珀移动通信有限公司 Video monitoring method and video monitoring terminal
CN103347156B (en) * 2013-06-24 2016-03-30 清华大学 A kind of method based on infrared camera data estimation wild animal population quantity
CN105404832B (en) * 2014-08-01 2019-08-16 优视科技有限公司 The safe handling method and device of mobile terminal
CN104573811A (en) * 2015-01-08 2015-04-29 杭州天迈网络有限公司 Pedestrian flow counting method based on infrared image and color image fusion
CN105404850B (en) * 2015-10-21 2019-03-05 中南大学 Station flow monitoring system for passenger, station passenger flow monitoring method and station passenger flow managing and control system, station passenger flow management-control method
CN106845344B (en) * 2016-12-15 2019-10-25 重庆凯泽科技股份有限公司 Demographics' method and device
CN106650695A (en) * 2016-12-30 2017-05-10 苏州万店掌网络科技有限公司 Video analysis technology-based people flow tracking statistics system
CN108833840A (en) * 2018-06-28 2018-11-16 苏州万店掌网络科技有限公司 A kind of shops's management inspection server

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101835032A (en) * 2010-04-27 2010-09-15 沈阳瑗玛信息技术有限公司 Multi-camera crowd-gathered message statistic device and method

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101835032A (en) * 2010-04-27 2010-09-15 沈阳瑗玛信息技术有限公司 Multi-camera crowd-gathered message statistic device and method

Also Published As

Publication number Publication date
CN102750539A (en) 2012-10-24

Similar Documents

Publication Publication Date Title
CN104303193B (en) Target classification based on cluster
US20190080461A9 (en) Neighborhood alert mode for triggering multi-device recording, multi-camera motion tracking, and multi-camera event stitching for audio/video recording and communication devices
CN102622588B (en) Dual-certification face anti-counterfeit method and device
CN104482715B (en) The management method of food and refrigerator in refrigerator
CN204463162U (en) A kind of intelligent face recognition system
RU2509355C2 (en) Apparatus and method of classifying movement of objects in monitoring zone
CN203588285U (en) Examinee identity identification device
CN103839346B (en) A kind of intelligent door and window anti-intrusion device and system, intelligent access control system
CN101587606B (en) Method and system for detecting staff flow in exhibition venue
CN104504362A (en) Face detection method based on convolutional neural network
US20110231419A1 (en) Systems, methods and articles for video analysis reporting
CN105934051B (en) Interactive classroom lighting intelligence control system and control method
CN103677267A (en) Mobile terminal and awakening method and device thereof
CN101383000B (en) Information processing apparatus, and information processing method
CN103106397B (en) Human face in-vivo detection method based on bright pupil effect
CN104268963A (en) Intelligent door lock system, intelligent door lock and intelligent alarm door
CN104091176B (en) Portrait comparison application technology in video
CN101552910B (en) Remnant detection device based on comprehensive computer vision
CN102157023B (en) Information system for self-help declaring and supervising of entry/exit passengers
CN101999888B (en) Epidemic preventing and controlling system for detecting and searching people with abnormal temperatures
CN101577812B (en) Method and system for post monitoring
CN104408406B (en) Personnel based on frame difference method and background subtraction leave the post detection method
CN103246869A (en) Crime monitoring method based on face recognition technology and behavior and sound recognition
CN106503687B (en) Merge the monitor video system for identifying figures and its method of face multi-angle feature
CN105913037A (en) Face identification and radio frequency identification based monitoring and tracking system

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model