CN103576840A - Stereoscopic vision based gesture body-sense control system - Google Patents
Stereoscopic vision based gesture body-sense control system Download PDFInfo
- Publication number
- CN103576840A CN103576840A CN201210257560.6A CN201210257560A CN103576840A CN 103576840 A CN103576840 A CN 103576840A CN 201210257560 A CN201210257560 A CN 201210257560A CN 103576840 A CN103576840 A CN 103576840A
- Authority
- CN
- China
- Prior art keywords
- image
- gesture body
- unit
- gesture
- body sense
- 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
Images
Abstract
The invention provides a stereoscopic vision based gesture body-sense control system. The system comprises a binocular video probe (1), a gesture body-sense recognition unit (2) and a control unit (3). The binocular video probe (1) collects gesture body-sense actions and outputs images to the gesture body-sense recognition unit (2). The gesture body-sense recognition unit (2) detects and recognizes the gesture body-sense actions and output recognition result to the control unit (3) which converts the gesture body-sense actions into control instructions to achieve corresponding operations. According to the system, the binocular video probe is used for obtaining depth information of the gesture body-sense actions, the Hu moment unchangeable characteristic and the depth information of the gesture body-sense actions are used, and a mature SVM classifier is used for achieving recognition and classification matching of the dynamic gesture body-sense actions.
Description
Technical field
The present invention relates to human-computer interaction technique field, field of video image processing, specifically the gesture body sense control system based on stereoscopic vision.
Background technology
Human-computer interaction technology mainly contains the gesture identification based on data glove at present, the gesture identification based on touch-screen, the gesture identification based on acceleration transducer, the gesture identification based on ultrasonic sensor and the gesture identification based on vision.Advantages such as the method based on data glove has quick and precisely, but the method requires user to dress complicated data glove and position tracker, does not meet the requirement of nature man-machine interaction, and data glove is expensive, is not suitable for large-scale popularization.Gesture input and identification based on touch-screen are very directly perceived, but this method needs user's contact screen, and by contrast, non-contacting mode can provide more natural man-machine interaction.The input of gesture based on acceleration transducer, user need to hold input equipment, and this input is easy to be interfered, and needs user to make cooperation.Gesture identification based on ultrasonic sensor exists such as the problem such as being easily disturbed, speed is slow.Gesture identification based on vision, has simple operation, the natural advantage of man-machine interaction.But present case is, monocular video cannot be found range and locate, and can only identify static gesture, and user can not self-defined gesture, and implanted chip equipment causes high cost, the defect such as cannot transplant, so that be difficult to commercialization penetration and promotion.
Summary of the invention
At least one object of the present invention is to provide a gesture body sense control system based on stereoscopic vision, to solve the technical matterss such as range finding location, the moved work of User Defined gesture body, many gestures body sense action recognition, dynamic gesture body sense action recognition, embedded independent control.
The new technical scheme that the above-mentioned technical matters that the present invention will solve adopts is achieved in that
A gesture body sense control system based on stereoscopic vision, comprising: binocular video probe, gesture body sense recognition unit and control module.Wherein, binocular video probe gathers gesture body and moves work, outputs image to gesture body sense recognition unit; The other unit inspection of gesture body perception and identification gesture body move to be done, and output recognition result is to control module; And the emotion of control module conversion gesture body is as steering order, realize respective operations.
Preferably, binocular video probe can three-dimensionally gather static gesture body and moves work and catch dynamic gesture body sense track, obtains the depth information of gesture body sense operating distance binocular video probe, carries out three-dimensional imaging, output imaging results.
Preferably, gesture body sense recognition unit is for supporting the graphics interface unit of hyperchannel I/O, comprise: image enhancing unit, image cutting unit, image detecting element, image identification unit, image data base, with image setting unit, the image that image enhancing unit can gather binocular video probe strengthens, image cutting unit can move the gesture body strengthening through image enhancing unit to do to split from general image, image detecting element can move and do to screen the gesture body being partitioned into through image cutting unit, extract characteristic action, the gesture body sense characteristic action that image identification unit is extracted image detecting element mates with the gesture body sense characteristic action prestoring in image data base, output matching result, the gesture body sense characteristic action that image setting unit extracts image detecting element deposits image data base in.
Preferably, control module comprises instruction converting unit and instruction database, and instruction converting unit is converted to steering order by the matching result of image identification unit output, and the steering order prestoring in and instruction database mates, and carries out respective operations.
Preferably, the communication between binocular video probe and gesture body sense recognition unit is connected at least one in following mode: TCP/IP, 2G, 3G, 4G, Wifi, Zigbee and W1an.
Preferably, the communication between gesture body sense recognition unit and control module is connected at least one in following mode: TCP/IP, 2G, 3G, 4G, Wifi, Zigbee and Wlan.
Technique effect of the present invention is:
Employing binocular video probe, obtains gesture body and moves the depth information of doing, and utilizes the moved Hu square invariant features of doing of gesture body and depth information thereof as training sample, utilizes ripe svm classifier device to realize dynamic gesture body and moves identification and the classification and matching of doing.Adopt specific gesture body to move and open and closing control system, prevent the maloperation of carrying out in unwanted situation.Adopt image enhancement technique, for the image under complex background and different illumination conditions, strengthen, with the gesture body of realizing under complex background and different illumination conditions, move extraction and the detection of doing.Adopt image Segmentation Technology and gesture priori, gesture body in outdoor scene is moved and does detect, cut apart and identify, realize many gestures body sense action control.Adopt on-line study technology, realize User Defined gesture body and move input and the management of doing.Adopt modern communication technology, realize the communication between gesture body sense recognition unit and control module, solve implanted chip equipment and cause the defect that cannot transplant.
Accompanying drawing explanation
Fig. 1: the Organization Chart that is the gesture body sense control system based on stereoscopic vision.
Fig. 2: the process flow diagram that is the gesture body sense control system based on stereoscopic vision.
Label declaration in schematic diagram:
1 binocular video probe, 2 gesture body sense recognition units, 3 control modules, 4 image enhancing unit, 5 image cutting units, 6 image detecting elements, 7 image identification unit, 8 image data bases, 9 image setting unit, 10 instruction converting unit, 11 instruction databases.
Embodiment
Below in conjunction with accompanying drawing, to the gesture body sense control system based on stereoscopic vision of the present invention, be that realization solves the optimal way that range finding location, User Defined gesture body moves the technical scheme that the technical matterss such as works, many gestures body sense action recognition, dynamic gesture body sense action recognition, embedded independently control take and is described further:
Referring to Fig. 1 and 2, the invention provides a kind of gesture body sense control system based on stereoscopic vision, comprising: binocular video probe 1, gesture body sense recognition unit 2 and control module 3; Wherein, binocular video probe 1 gathers gesture body and moves work, outputs image to gesture body sense recognition unit 2; Gesture body sense recognition unit 2 detection and Identification gesture bodies move to be done, and output recognition result is to control module 3; And control module 3 conversion gesture body emotions are as steering order, realize respective operations.
Particularly, binocular video probe 1 can three-dimensionally gather static gesture body and moves work and catch dynamic gesture body sense track, obtains the depth information of gesture body sense operating distance binocular video probe 1, carries out three-dimensional imaging, output imaging results.
Particularly, gesture body sense recognition unit 2, for supporting the graphics interface unit of hyperchannel I/O, comprising: image enhancing unit 4, image cutting unit 5, image detecting element 6, image identification unit 7, image data base 8 and image setting unit 9, wherein, image enhancing unit 4 can strengthen the pop one's head in image of 1 collection of binocular video, image cutting unit 5 can move the gesture body strengthening through image enhancing unit 4 to do to split from general image, image detecting element 6 can move and do to screen the gesture body being partitioned into through image cutting unit 5, extract characteristic action, the gesture body sense characteristic action that image identification unit 7 is extracted image detecting element 6 mates with the gesture body sense characteristic action prestoring in image data base 8, output matching result, the gesture body sense characteristic action that image setting unit 9 extracts image detecting element 6 deposits image data base 8 in.
Particularly, control module 3 comprises instruction converting unit 10 and instruction database 11; Wherein, instruction converting unit 10 is converted to steering order by the matching result of image identification unit 7 output, and the steering order prestoring in and instruction database 11 mates, and carries out respective operations.
Preferably, the communication between binocular video probe 1 and gesture body sense recognition unit 2 is connected at least one in following mode: TCP/IP, 2G, 3G, 4G, Wifi, Zigbee and Wlan.
Preferably, the communication between gesture body sense recognition unit 2 and control module 3 is connected at least one in following mode: TCP/IP, 2G, 3G, 4G, Wifi, Zigbee and Wlan.
It is below the preferred operations flow process of a kind of gesture body sense control system based on stereoscopic vision provided by the invention.
Step 1: binocular video probe (1) is three-dimensional to be gathered the moved work of static gesture body and catches dynamic gesture body sense track, obtain the depth information of gesture body sense operating distance binocular video probe (1), carry out three-dimensional imaging, output imaging results, under good illumination condition, forward step 3 to;
Step 2: binocular video probe (1) is three-dimensional to be gathered the moved work of static gesture body and catches dynamic gesture body sense track, obtain the depth information of gesture body sense operating distance binocular video probe (1), carry out three-dimensional imaging, output imaging results, under bad illumination condition, forward step 4 to;
Step 3 image cutting unit (5) moves the gesture body strengthening through image enhancing unit (4) to do to split from general image, image detecting element (6) moves and does to screen the gesture body being partitioned into through image cutting unit (5), extract characteristic action, the gesture body sense characteristic action that image identification unit (7) is extracted image detecting element (6) mates with the gesture body sense characteristic action prestoring in image data base (8), output matching result, forwards step 5 to;
Step 4: the image that image enhancing unit (4) gathers binocular video probe (1) strengthens, and forwards step 3 to;
Step 5: meet breakdown action, open control module (3), enter gesture body sense controlling unit, realize dynamic gesture body sense action recognition and many gestures body sense action recognition, forward step 7 to;
Step 6: meet breakdown action, open control module (3), enter User Defined gesture body sense action link, forward step 8 to;
Step 7: instruction converting unit (10) is converted to steering order by the matching result of image identification unit (7) output, and the steering order prestoring in and instruction database (11) is mated, carry out respective operations, operated backed off after random control system;
Step 8: user inputs certain gestures body and moves work, the gesture body sense characteristic action that image setting unit (9) extracts image detecting element (6) deposits image data base (8) in, realize action coupling, operated backed off after random control system, again enter available.
Claims (6)
1. the gesture body sense control system based on stereoscopic vision, comprising:
Binocular video probe (1), gesture body sense recognition unit (2) and control module (3);
Wherein, binocular video probe (1) gathers gesture body and moves work, outputs image to gesture body sense recognition unit (2); Gesture body sense recognition unit (2) detection and Identification gesture body moves to be done, and output recognition result is to control module (3); And the emotion of control module (3) conversion gesture body is as steering order, realize respective operations.
2. system according to claim 1, binocular video probe (1) described in it is characterized in that can three-dimensionally gather static gesture body and moves work and catch dynamic gesture body sense track, obtain the depth information of gesture body sense operating distance binocular video probe (1), carry out three-dimensional imaging, output imaging results.
3. system according to claim 1, it is characterized in that described gesture body sense recognition unit (2), for supporting the graphics interface unit of hyperchannel I/O, comprising: image enhancing unit (4), image cutting unit (5), image detecting element (6), image identification unit (7), image data base (8) and image setting unit (9);
Wherein, image enhancing unit (4) image that (1) gather of can popping one's head in to binocular video strengthens, image cutting unit (5) can move the gesture body strengthening through image enhancing unit (4) to do to split from general image, image detecting element (6) can move and do to screen the gesture body being partitioned into through image cutting unit (5), extract characteristic action, the gesture body sense characteristic action that image identification unit (7) is extracted image detecting element (6) mates with the gesture body sense characteristic action prestoring in image data base (8), output matching result, the gesture body sense characteristic action that image setting unit (9) extracts image detecting element (6) deposits image data base (8) in.
4. system according to claim 1, is characterized in that described control module (3) comprises instruction converting unit (10) and instruction database (11);
Wherein, instruction converting unit (10) is converted to steering order by the matching result of image identification unit (7) output, and the steering order prestoring in and instruction database (11) mates, and carries out respective operations.
5. system according to claim 1, is characterized in that described binocular video probe (1) and the communication between gesture body sense recognition unit (2) are connected at least one in following mode: TCP/IP, 2G, 3G, 4G, Wifi, Zigbee and Wlan.
6. system according to claim 1, is characterized in that described gesture body sense recognition unit (2) and the communication between control module (3) are connected at least one in following mode: TCP/IP, 2G, 3G, 4G, Wifi, Zigbee and Wlan.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210257560.6A CN103576840A (en) | 2012-07-24 | 2012-07-24 | Stereoscopic vision based gesture body-sense control system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210257560.6A CN103576840A (en) | 2012-07-24 | 2012-07-24 | Stereoscopic vision based gesture body-sense control system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103576840A true CN103576840A (en) | 2014-02-12 |
Family
ID=50048801
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210257560.6A Pending CN103576840A (en) | 2012-07-24 | 2012-07-24 | Stereoscopic vision based gesture body-sense control system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103576840A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104808799A (en) * | 2015-05-20 | 2015-07-29 | 成都通甲优博科技有限责任公司 | Unmanned aerial vehicle capable of indentifying gesture and identifying method thereof |
CN105677032A (en) * | 2016-01-04 | 2016-06-15 | 广州华欣电子科技有限公司 | Control method and device based on depth sensing equipment |
CN105892633A (en) * | 2015-11-18 | 2016-08-24 | 乐视致新电子科技(天津)有限公司 | Gesture identification method and virtual reality display output device |
CN106205237A (en) * | 2016-08-31 | 2016-12-07 | 律世刚 | Based on movement response and the training method of the second mother tongue of drawing reaction and device |
CN106648112A (en) * | 2017-01-07 | 2017-05-10 | 武克易 | Somatosensory action recognition method |
CN106708266A (en) * | 2016-12-23 | 2017-05-24 | 武汉市马里欧网络有限公司 | AR action correction projection method and system based on binocular gesture recognition |
CN106861179A (en) * | 2016-12-27 | 2017-06-20 | 苏州春禄电子科技有限公司 | A kind of mobile phone games control system based on body-sensing technology |
CN106873764A (en) * | 2017-04-14 | 2017-06-20 | 苏州春禄电子科技有限公司 | A kind of mobile phone gesture input systems based on motion sensing control system |
US9704251B2 (en) | 2014-10-11 | 2017-07-11 | Boe Technology Group Co., Ltd. | Depth determination method, depth determination device and electronic device |
WO2017118075A1 (en) * | 2016-01-04 | 2017-07-13 | 京东方科技集团股份有限公司 | Human-machine interaction system, method and apparatus |
CN107450717A (en) * | 2016-05-31 | 2017-12-08 | 联想(北京)有限公司 | A kind of information processing method and Wearable |
CN107920203A (en) * | 2017-11-23 | 2018-04-17 | 乐蜜有限公司 | Image-pickup method, device and electronic equipment |
CN110850982A (en) * | 2019-11-11 | 2020-02-28 | 南方科技大学 | AR-based human-computer interaction learning method, system, device and storage medium |
CN111491066A (en) * | 2020-03-14 | 2020-08-04 | 武汉中观自动化科技有限公司 | Tracking type scanning device and method supporting gesture control |
-
2012
- 2012-07-24 CN CN201210257560.6A patent/CN103576840A/en active Pending
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9704251B2 (en) | 2014-10-11 | 2017-07-11 | Boe Technology Group Co., Ltd. | Depth determination method, depth determination device and electronic device |
CN104808799A (en) * | 2015-05-20 | 2015-07-29 | 成都通甲优博科技有限责任公司 | Unmanned aerial vehicle capable of indentifying gesture and identifying method thereof |
WO2017084319A1 (en) * | 2015-11-18 | 2017-05-26 | 乐视控股(北京)有限公司 | Gesture recognition method and virtual reality display output device |
CN105892633A (en) * | 2015-11-18 | 2016-08-24 | 乐视致新电子科技(天津)有限公司 | Gesture identification method and virtual reality display output device |
US10585488B2 (en) | 2016-01-04 | 2020-03-10 | Boe Technology Group Co., Ltd. | System, method, and apparatus for man-machine interaction |
CN105677032A (en) * | 2016-01-04 | 2016-06-15 | 广州华欣电子科技有限公司 | Control method and device based on depth sensing equipment |
WO2017118075A1 (en) * | 2016-01-04 | 2017-07-13 | 京东方科技集团股份有限公司 | Human-machine interaction system, method and apparatus |
CN105677032B (en) * | 2016-01-04 | 2018-10-26 | 广州华欣电子科技有限公司 | Control method and device based on depth sensing equipment |
CN107450717A (en) * | 2016-05-31 | 2017-12-08 | 联想(北京)有限公司 | A kind of information processing method and Wearable |
CN107450717B (en) * | 2016-05-31 | 2021-05-18 | 联想(北京)有限公司 | Information processing method and wearable device |
CN106205237A (en) * | 2016-08-31 | 2016-12-07 | 律世刚 | Based on movement response and the training method of the second mother tongue of drawing reaction and device |
CN106708266A (en) * | 2016-12-23 | 2017-05-24 | 武汉市马里欧网络有限公司 | AR action correction projection method and system based on binocular gesture recognition |
CN106861179A (en) * | 2016-12-27 | 2017-06-20 | 苏州春禄电子科技有限公司 | A kind of mobile phone games control system based on body-sensing technology |
CN106648112A (en) * | 2017-01-07 | 2017-05-10 | 武克易 | Somatosensory action recognition method |
CN106873764A (en) * | 2017-04-14 | 2017-06-20 | 苏州春禄电子科技有限公司 | A kind of mobile phone gesture input systems based on motion sensing control system |
CN107920203A (en) * | 2017-11-23 | 2018-04-17 | 乐蜜有限公司 | Image-pickup method, device and electronic equipment |
CN110850982A (en) * | 2019-11-11 | 2020-02-28 | 南方科技大学 | AR-based human-computer interaction learning method, system, device and storage medium |
CN110850982B (en) * | 2019-11-11 | 2024-03-29 | 南方科技大学 | AR-based man-machine interaction learning method, system, equipment and storage medium |
CN111491066A (en) * | 2020-03-14 | 2020-08-04 | 武汉中观自动化科技有限公司 | Tracking type scanning device and method supporting gesture control |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103576840A (en) | Stereoscopic vision based gesture body-sense control system | |
EP3961485A1 (en) | Image processing method, apparatus and device, and storage medium | |
CN102854983B (en) | A kind of man-machine interaction method based on gesture identification | |
Gu et al. | Human gesture recognition through a kinect sensor | |
Hasan et al. | Brightness factor matching for gesture recognition system using scaled normalization | |
CN103530892B (en) | A kind of both hands tracking based on Kinect sensor and device | |
CN102103409A (en) | Man-machine interaction method and device based on motion trail identification | |
CN102945362B (en) | The sensor synergism gesture identification method merged based on isomeric data and system | |
CN111104820A (en) | Gesture recognition method based on deep learning | |
CN103514437B (en) | A kind of three-dimension gesture identifying device and three-dimensional gesture recognition method | |
CN102945078A (en) | Human-computer interaction equipment and human-computer interaction method | |
JP2017505965A (en) | Real-time 3D gesture recognition and tracking system for mobile devices | |
CN106814853A (en) | A kind of eye control tracking based on machine learning | |
CN105913034B (en) | Vehicle identification method and device and vehicle | |
CN205721986U (en) | A kind of cable visual defects detection and character recognition device | |
CN103713755A (en) | Touch recognizing device and recognizing method | |
CN109189219A (en) | The implementation method of contactless virtual mouse based on gesture identification | |
CN107390881A (en) | A kind of gestural control method | |
CN202584048U (en) | Smart mouse based on DSP image location and voice recognition | |
CN109634408A (en) | A kind of extended method of Hololens gesture identification | |
Howe et al. | Comparison of hand segmentation methodologies for hand gesture recognition | |
Rupanagudi et al. | A novel and secure methodology for keyless ignition and controlling an automobile using air gestures | |
Ying et al. | Fingertip detection and tracking using 2D and 3D information | |
Dadiz et al. | Go-Mo (Go-Motion): An android mobile application detecting motion gestures for generating basic mobile phone commands utilizing KLT algorithm | |
CN114296543A (en) | Fingertip force detection and gesture recognition intelligent interaction system and intelligent ring |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
DD01 | Delivery of document by public notice |
Addressee: SHANGHAI CHENXU INFORMATION TECHNOLOGY CO., LTD. Document name: Notification of Publication of the Application for Invention |
|
DD01 | Delivery of document by public notice |
Addressee: SHANGHAI CHENXU INFORMATION TECHNOLOGY CO., LTD. Document name: Notification that Application Deemed to be Withdrawn |
|
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140212 |