CN1662060A - System of identifying image of human bodies - Google Patents
System of identifying image of human bodies Download PDFInfo
- Publication number
- CN1662060A CN1662060A CN 200410006024 CN200410006024A CN1662060A CN 1662060 A CN1662060 A CN 1662060A CN 200410006024 CN200410006024 CN 200410006024 CN 200410006024 A CN200410006024 A CN 200410006024A CN 1662060 A CN1662060 A CN 1662060A
- Authority
- CN
- China
- Prior art keywords
- judging unit
- people
- portrait
- image
- recognition system
- 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
Images
Landscapes
- Closed-Circuit Television Systems (AREA)
- Burglar Alarm Systems (AREA)
Abstract
The system includes an image picking up unit and a judging unit. The image picking up unit can be CCD or CMOS camera, and images picked up are transferred to the judging unit. The said judging unit can be a microprocessor or computer. At least a person's image permitted to pass through is setup in memory of the judging unit in advance. Based on image signals picked up by the picking up in series, the judging unit determines whether there is a moving body. The said moving body is compared with person's image stored in memory in order to determine whether the moving body is a person and moving direction and location. The method determines whether the person is going to enter to room or exit from the room. If the person is allowed passing through, then signal for opening the door is sent out.
Description
Technical field
The present invention is a kind of body image recognition system, refers to especially a kind ofly can judge whether the recognition system that door is opened according to continuous picked image signal.
Technical background
Press, the recognition system of general automatically-controlled door or gate inhibition's room entry/exit management is to be widely used in various occasions, such as: even shop, department store, dwelling house, office, save antitheft, self-service bank, military affairs toy etc. from damage.The technology of this recognition system is broadly divided into initiatively two kinds of checking systems and passive checking system, wherein this active checking system must initiatively be launched a signal (as: laser signal, infra-red signal, ultrasonic waves signal, radar signal etc.) to an environment to be measured by measurement person, and by the device receiver, the signal that reception is returned by Ambient to be measured is analyzed (information such as signal power or time difference), with push away the information (as: orientation, size, distance) of determinand, give suitable differentiation action again.Passive checking system then can be by the infrared wave launched because of body temperature of detecting human body etc., push away the information of determinand, give suitable differentiation action again.
Yet this active checking system is provided with initiatively signal emission and the device that receives, and not only increases system complexity and cost, also has to measure inaccurate and cause the disappearance of result's erroneous judgement, and considering of power consumption arranged.The technology of the passive checking system of infrared ray then is subject to ambient temperature, human body and is capped factor such as rate and influences the interpretation situation, and expensive equipment cost is not to be fit to general application scenario yet.
Summary of the invention
Because the disappearance of aforementioned located by prior art, the inventor is manufacturing experience and the technology accumulation of being engaged in image recognition system and electronic installation according to it, method at the various solutions of the concentrated research of above-mentioned disappearance, after through constantly research, experiment and improvement, finally development and Design goes out a kind of environmental images that is captured with the GENERAL TYPE video camera of the present invention, go out wherein people information via the interpretation of image identification program, as the body image recognition system of the keying of door, many disappearances that its prior art of forgoing is derived.
A purpose of the present invention, provide a kind of body image recognition system of tool behavior interpretation function, this system mainly comprises an image acquisition unit and a judging unit, wherein, this image acquisition unit can be a video camera, it is sent to judging unit with picked image, and this judging unit can be a microprocessor or computer, be set with at least one portrait profile in the memory body of this judging unit in advance, this judging unit can be according to the continuous picked image signal of this image acquisition unit, judged whether mobile object, and the portrait profile in this mobile object and the memory body compared, whether be the people to judge this mobile object, again according to the behavior pattern of its position and continuous moving direction, judge whether this people desires to enter (going out) indoor (outward), whether this people's portrait and the portrait in the memory body are compared if necessary, serve as to permit current people to judge this people, if should artificially permit current people again, then send enabling signal door is opened, make this people be able to by this door.
By technical scheme of the present invention, will bring following beneficial effect:
1. significantly improve the frequent false signal action disappearance (opening the infrared induction door) of the equipment of commonly using as doggie.
2. combine with video camera (making video recording), can increase understanding, be beneficial to the ex-post analysis interpretation of emergency case environmental information.
3. can satisfy in conjunction with other identification mechanism (as: human face recognition) and harsh save demand from damage or move as criminal and investigate application.
4. equipment requirements is simple, with low cost, even if easy CMOS (Complementary Metal Oxide Semiconductor) conductor (CMOS) video camera of arranging in pairs or groups also can be operated.
5. the algorithm implementation is full of elasticity, can revise fast and use at different demands.
Description of drawings
Fig. 1 is a block schematic diagram of the present invention;
Fig. 2 is an action flow chart of the present invention;
Fig. 3 is that the present invention judges one of schematic diagram of image;
Fig. 4 is two of the present invention's schematic diagram of judging image.
Controlled device 13
Embodiment
The present invention is a kind of body image recognition system, see also shown in Figure 1, this system mainly comprises an image acquisition unit 11 and a judging unit 12, wherein, this image acquisition unit 11 can be a video camera (as: video camera of charge coupled cell (CCD), the video camera of CMOS (Complementary Metal Oxide Semiconductor) conductor (CMOS) etc.), it is sent to judging unit 12 with picked image, and this judging unit 12 can be a microprocessor, personal computer or digital signal processor, store at least one in the memory body 121 of this judging unit 12 in advance and permit current portrait (in the present invention, this portrait comprises portrait external form and specific people's portrait feature etc.), make this judging unit 12 differentiate this image acquisition unit 11 picked image according to the portrait in this memory body 121, this judging unit 12 also is electrically connected with a controlled device 13 (as: automatically-controlled door).
The composition of the above-mentioned member of mat, see also shown in Figure 1 again, this image acquisition unit 11 can constantly be sent to picked image this judging unit 12, make this judging unit 12 can be according to these image acquisition unit 11 continuous picked image signals, judged whether mobile object, and the portrait profile in this mobile object and the memory body 121 compared, to judge whether this mobile object is the people, comply with the behavior pattern of its position and continuous moving direction again, judge whether this people desires to enter (going out) indoor (outward), succeed and again this people's portrait and the portrait in the memory body 121 are compared, to judge whether this people serves as to permit current people, so, if this people permits current people, then send a start signal to this controlled device 13, make this controlled device 13 open, and then this people is able to by this controlled device 13 (for example: send an enabling signal door is opened, make this people be able to by this door).
See also shown in Fig. 1,2, when this system is used, its judging unit 12 will be handled according to following step:
(201) receive the image that transmits by this image acquisition unit 11;
Does (202) the preceding image of comparing these images have different part with the back image? if have, continue following step (203); If not, carry out step (201);
(203) calculate the vector of forward and backward image according to these preceding eiconometers;
(204) do you judge that these preceding images and back image are mobile object? if continue following step (205); If not, carry out step (201);
Is (205) the portrait profile in these images and the memory body 121 being compared, the people to judge these images? if continue following step 206; Otherwise, carry out step (201);
(206) do you judge that this vector is for desiring to enter (going out) indoor (outward)? if continue following step (207); Otherwise, carry out step (201);
Whether (207) this people's portrait and the portrait in the memory body 121 are compared, serve as to permit current people, and distinguish the direction that this people moves continuously to judge this people, move towards the door center? if continue following step (208); Otherwise, carry out step (201);
(208) send an enabling signal door is opened, make this people be able to, finish again by this door.
In the present invention, when this judging unit 12 is judged these image acquisition unit 11 picked image signals, be to carry out according to two stages, at first take out the difference part of the forward and backward image of these signal of video signal, can isolate and move part and static background, according to the next moving object position of obtaining, relatively the central point of this two mobile object changes again, can draw the direction and the speed of this movement of objects.According to the moving direction (whether mobile object is by left and right, upward mobile to image center point) and the velocity variations of this mobile object, can learn whether this mobile object desires to enter (going out) indoor (outward).Carry out the identification of portrait then, the identification of portrait comprises rejects false signal and colour of skin identification, features identification.Whether final decision is opened door, if then add portrait comparison function as entrance guard management.According to this people advance direction position and translational speed, calculate and when decision begins door is opened.Before beginning that in decision door opened time point, this people turns (changing mind), still can cancel the action of opening the door.
See also shown in Figure 3, if target object is at image I
tIn coordinate be (i, in the time of j), can be at image I
T+ Δ tLast every bit coordinate searches one one by one, and (square of M * M) finds ∑
M, n=1; M(| I
T+ Δ t(the i+m+ Δ, j+n)-I
t(i+m, j+n) |)/64 get coordinate (i+ Δ, 1) when minimum values take place, and the pixel difference is Δ.
I wherein
t(i+m, j+n) (i+m, j+n) brightness value of image of coordinate points in the image of representative when the time is t.So, can calculate the motion vector of each pixel (pixel) in whole the image, select the zone of same motion vector, can doubtful candidate personage be found out by circle.See also shown in Figure 4ly, analyze the motion vector (level and vertical component) calculate again, predict and assess the moving direction and the speed of this object, differentiate its movement locus tendency, with its behavior differentiation as corresponding switch motion foundation.
Whether in addition, at least one portrait that stores in advance in the memory body 121 of this judging unit 12 can be a portrait external form, being that people's discriminating is used as this mobile object.
Whether at least one portrait that stores in advance in the memory body 121 of this judging unit 12 also can be a specific people portrait feature, being that specific people's discriminating is used as this mobile object.
Beneficial functional of the present invention is as described below:
1. significantly improve the frequent false signal start disappearance (opening the infrared induction door) of the equipment of commonly using as doggie.
2. combine with video camera (making video recording), can increase understanding, be beneficial to the ex-post analysis interpretation of emergency case environmental information.
3. can satisfy in conjunction with other identification mechanism (as: human face recognition) and harsh save demand from damage or move as criminal and investigate application.
4. equipment requirements is simple, with low cost, even if easy CMOS (Complementary Metal Oxide Semiconductor) conductor (CMOS) video camera of arranging in pairs or groups also can be operated.
5. the algorithm implementation is full of elasticity, can revise fast and use at different demands.
The above only is the best specific embodiment of the present invention, however structural feature of the present invention be not limited thereto, anyly be familiar with this skill person in field of the present invention, can think easily and variation or modification, all can be encompassed in the scope of patent protection of this case.
Claims (9)
1. body image recognition system is characterized in that comprising:
One image acquisition unit, it is sent to judging unit with picked image;
One judging unit stores at least one portrait in advance in the memory body of this judging unit, make this judging unit differentiate this image acquisition unit picked image according to the portrait in the memory body;
This image acquisition unit can constantly be sent to picked image this judging unit, make this judging unit can be according to the continuous picked image signal of this image acquisition unit, judged whether mobile object, and the portrait in this mobile object and the memory body compared, to judge whether this mobile object is the people, change to judge its behavior pattern according to the continuous moving direction of this mobile object again, again according to result of determination, the output control signal.
2. body image recognition system as claimed in claim 1, it is characterized in that: this judging unit and a controlled device are electrically connected, it is man-hour that this judgment unit judges goes out this mobile object, direction and the position that can move according to this people again, judge whether this people desires to enter (going out) indoor (outward), again this people's portrait and the portrait in the memory body are compared, to judge whether this people serves as to permit current people, so, if this people permits current people, then send a start signal to this controlled device, make this controlled device open.
3. body image recognition system as claimed in claim 1 is characterized in that: at least one portrait that stores in advance in the memory body of this judging unit can be a portrait external form, whether being that people's discriminating is used as this mobile object.
4. body image recognition system as claimed in claim 1 is characterized in that: at least one portrait that stores in advance in the memory body of this judging unit can be a specific portrait feature, whether being that specific people's discriminating is used as this mobile object.
5. body image recognition system as claimed in claim 1 is characterized in that: this image acquisition unit can be the video camera of a charge coupled cell (CCD).
6. body image recognition system as claimed in claim 1 is characterized in that: this image acquisition unit can be the video camera of a CMOS (Complementary Metal Oxide Semiconductor) conductor (CMOS).
7. body image recognition system as claimed in claim 2 is characterized in that: this judging unit can be a microprocessor.
8. body image recognition system as claimed in claim 2 is characterized in that: this judging unit can be a personal computer.
9. body image recognition system as claimed in claim 2 is characterized in that: this controlled device can be an automatically-controlled door.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNB2004100060244A CN100385947C (en) | 2004-02-25 | 2004-02-25 | System of identifying image of human bodies |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNB2004100060244A CN100385947C (en) | 2004-02-25 | 2004-02-25 | System of identifying image of human bodies |
Publications (2)
Publication Number | Publication Date |
---|---|
CN1662060A true CN1662060A (en) | 2005-08-31 |
CN100385947C CN100385947C (en) | 2008-04-30 |
Family
ID=35011097
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNB2004100060244A Expired - Fee Related CN100385947C (en) | 2004-02-25 | 2004-02-25 | System of identifying image of human bodies |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN100385947C (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101975009A (en) * | 2010-10-29 | 2011-02-16 | 无锡中星微电子有限公司 | Automatic door control device and method |
CN103123768A (en) * | 2011-11-18 | 2013-05-29 | 鸿富锦精密工业(深圳)有限公司 | Rotary advertisement board |
CN104899994A (en) * | 2014-03-05 | 2015-09-09 | 昆山毅达科技电子有限公司 | Intelligent alarm system for fireproof escape door |
CN105931326A (en) * | 2016-04-15 | 2016-09-07 | 刘双文 | Access control device and method |
CN106600774A (en) * | 2016-12-07 | 2017-04-26 | 深圳市万睿智能科技有限公司 | Face identification gate inhibition control method provided with door opening detection area and system thereof |
CN106736824A (en) * | 2015-11-25 | 2017-05-31 | 亚太菁英股份有限公司 | Intelligent anticollision security system and apply its toolroom machine |
CN112880155A (en) * | 2021-02-07 | 2021-06-01 | 河北红岸基地科技有限公司 | Intelligent air conditioning equipment operation control method based on camera recognition |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1150373C (en) * | 2001-02-19 | 2004-05-19 | 长春当代信息产业集团有限公司 | Entrance guard system with human image recognition |
JP2003027796A (en) * | 2001-07-23 | 2003-01-29 | Oki Electric Ind Co Ltd | Entrance/leaving control system |
JP2003308303A (en) * | 2002-04-18 | 2003-10-31 | Toshiba Corp | Person authentication system, and passing control system |
CN2566379Y (en) * | 2002-09-06 | 2003-08-13 | 北京中星微电子有限公司 | Face identification type door control device |
-
2004
- 2004-02-25 CN CNB2004100060244A patent/CN100385947C/en not_active Expired - Fee Related
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101975009A (en) * | 2010-10-29 | 2011-02-16 | 无锡中星微电子有限公司 | Automatic door control device and method |
CN103123768A (en) * | 2011-11-18 | 2013-05-29 | 鸿富锦精密工业(深圳)有限公司 | Rotary advertisement board |
CN104899994A (en) * | 2014-03-05 | 2015-09-09 | 昆山毅达科技电子有限公司 | Intelligent alarm system for fireproof escape door |
CN104899994B (en) * | 2014-03-05 | 2017-10-03 | 昆山毅达科技电子有限公司 | The intelligent warning system of fire-proof escaping door |
CN106736824A (en) * | 2015-11-25 | 2017-05-31 | 亚太菁英股份有限公司 | Intelligent anticollision security system and apply its toolroom machine |
CN105931326A (en) * | 2016-04-15 | 2016-09-07 | 刘双文 | Access control device and method |
CN106600774A (en) * | 2016-12-07 | 2017-04-26 | 深圳市万睿智能科技有限公司 | Face identification gate inhibition control method provided with door opening detection area and system thereof |
CN112880155A (en) * | 2021-02-07 | 2021-06-01 | 河北红岸基地科技有限公司 | Intelligent air conditioning equipment operation control method based on camera recognition |
Also Published As
Publication number | Publication date |
---|---|
CN100385947C (en) | 2008-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xu et al. | A people counting system based on head-shoulder detection and tracking in surveillance video | |
CN1206149C (en) | System for safe-checking and transporting to passengers with elevator equipment, its running method and supplementary mounting method | |
CN105868689A (en) | Cascaded convolutional neural network based human face occlusion detection method | |
US10977826B1 (en) | Safety detection camera system for door closure | |
Lengvenis et al. | Application of computer vision systems for passenger counting in public transport | |
CN109101898A (en) | Public transport vehicle-mounted number real-time monitoring system and its monitoring method | |
KR101917783B1 (en) | Parking lot management system | |
US20090060292A1 (en) | Image input apparatus, and image input method | |
US20080298687A1 (en) | Human image recognition system | |
CN114973207A (en) | Road sign identification method based on target detection | |
CN1662060A (en) | System of identifying image of human bodies | |
US11030465B1 (en) | Method for analyzing number of people and system thereof | |
CN112784759A (en) | Elevator human detection identification method based on artificial intelligence similarity comparison | |
CN117037085A (en) | Vehicle identification and quantity statistics monitoring method based on improved YOLOv5 | |
KR101924643B1 (en) | Recognition system for vehicle | |
CN114359839A (en) | Method and system for identifying entrance of electric vehicle into elevator | |
CN113622786B (en) | Automatic door control method, system and equipment | |
CN116246402A (en) | Monitoring method and device | |
US11645847B2 (en) | Vehicle control system, vehicle control method, and recording medium in which vehicle control program is recorded | |
US20210250554A1 (en) | Vehicle control system, vehicle control method, and recording medium in which vehicle control program is recorded | |
CN103902990A (en) | Method for rapidly detecting facial features of driver based on Haar-like features | |
CN110097758A (en) | Information of vehicles output, storage method and device | |
Kumagai et al. | Improving Accuracy of Traffic Sign Detection Using Learning Method | |
Khantasak et al. | Parking violation detection system based on video processing | |
CN110633670A (en) | Method, system and device for prolonging green letter special for old people |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080430 Termination date: 20130225 |