CN106377265A - Behavior detection system based on depth image and eye movement watching information - Google Patents
Behavior detection system based on depth image and eye movement watching information Download PDFInfo
- Publication number
- CN106377265A CN106377265A CN201610837316.5A CN201610837316A CN106377265A CN 106377265 A CN106377265 A CN 106377265A CN 201610837316 A CN201610837316 A CN 201610837316A CN 106377265 A CN106377265 A CN 106377265A
- Authority
- CN
- China
- Prior art keywords
- detection terminal
- depth image
- eye
- control centre
- data
- 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.)
- Withdrawn
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1103—Detecting eye twinkling
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1123—Discriminating type of movement, e.g. walking or running
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Dentistry (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Physiology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Ophthalmology & Optometry (AREA)
- Alarm Systems (AREA)
Abstract
The invention relates to a behavior detection system based on a depth image and eye movement watching information. The behavior detection system comprises a control center, a plurality of detection terminals which are connected with the control center in a communication manner and are in network-like distribution, and a data acquisition terminal connected with the detection terminals in a communication manner, wherein the data acquisition terminal comprises a depth image sensor, an eye tracker sensor, a colored image sensor, a voice sensor and a thermal image sensor. The behavior detection system solves the problems that existing behavior detection system only has a single type of equipment and can detect singly in the prior art, and the behavior detection system disclosed by the invention is simple in structure and can be realized by structural adjustment by using existing equipment facilities, so that the transformation cost is low.
Description
Technical field
The invention belongs to behavioral value data acquisition technology field is and in particular to a kind of being moved based on depth image and eye is watched attentively
The behavioral value system of information.
Background technology
With artificial intelligence and mode identification technology development and a large amount of use, multi-modal data acquisition equipment increasingly increases
Many, accordingly, carry out Activity recognition using these data and cognitive technology becomes increasingly complex;Traditional behavioral value
Mainly obtain gray scale or the coloured image of people by photographic head, and carry out certain specific mesh using image procossing and analysis method
Identification, such work is little to the scope of behavioral value, and error is big, high cost, the information to behavioral value needs itself
Degrees of fusion is poor, has significant limitation.Based on this, for issuable various problems, those skilled in the art's design is opened
Send out the collecting device of various actions data.But above-mentioned equipment is often " autonomous operation ", the information gathering of most collection equipments
Single function, information gathering content is limited, with regard to a small amount of using effect of small range substantially, should for more complicated behavioral value
With still suffering from the problems such as cost, information fusion, operating error.
Content of the invention
It is an object of the invention to solving above-mentioned technical problem and providing a kind of moving based on depth image and eye to watch letter attentively
The behavioral value system of breath.
For achieving the above object, the present invention adopts the following technical scheme that:
A kind of the behavioral value system watching information attentively is moved based on depth image and eye,
Including control centre, also include with described control centre communication connection net distribution multiple detection terminals
And the data collection station with the communication connection of described detection terminal;Described data collection station include depth image sensor,
Eye tracker sensor, color image sensor, sound transducer, thermal imagery sensor.
Described detection terminal, the database server be connected including arithmetic server and with described arithmetic server.
Described detection terminal includes one-level detection terminal and secondary detection terminal, described one-level detection terminal respectively with institute
State data collection station, the transmission of number data is realized in control centre's communication connection;Described secondary detection terminal respectively with described data
The transmission of number data is realized in acquisition terminal, the communication connection of one-level detection terminal.
Described control centre includes machine learning server and the data base's clothes being connected with described machine learning server
Business device.
It is single that the present invention solves behavioral value system equipment in prior art, and detection simple question is it is achieved that depth
Image and eye move the stereopsis result watching information attentively, meanwhile, solve detection information ditch using the net distribution of detection terminal
Lead to not smooth problem it is achieved that the behavior cognition detection of integrated multi-modal data;Can be calculated by the machine learning of detection terminal
Method carries out pattern recognition, forms more accurate biological information, behavior characteristic information and eye and moves and watch information attentively;The present invention ties
Structure is simple, carries out structural adjustment using existing equipment facility and can achieve, improvement cost is low.
Brief description
Fig. 1 illustrates the structure arrangement signal that the present invention moves the behavioral value system watching information attentively based on depth image and eye
Figure.
Specific embodiment
Below, in conjunction with example, the substantive distinguishing features of the present invention and advantage are further described, but the present invention not office
It is limited to listed embodiment.
Shown in Figure 1, a kind of being moved based on depth image and eye is watched attentively the behavioral value system of information and is included control centre,
Also include with described control centre communication connection net distribution multiple detection terminals and with described detection terminal lead to
The data collection station that letter connects;Described data collection station includes depth image sensor, eye tracker sensor, coloured image
Sensor, sound transducer, thermal imagery sensor.
In the present invention, described detection terminal can adopt existing living things feature recognition equipment, various biological special to collection
Levy data to be identified processing, formed and the structuring number watching information attentively is moved based on biological information, Activity recognition information and eye
According to.
As an embodiment, in the present invention, described detection terminal can be including such as arithmetic server and with described
The database server that arithmetic server is connected, described arithmetic server and described data collection station communication connection.
It should be noted that, in the present invention, described arithmetic server adopts corresponding arithmetic server in prior art,
Mainly pattern recognition is carried out by existing corresponding machine learning algorithm, formed based on biological information, Activity recognition letter
Breath and eye move the structural data watching information attentively, then by formed structural data and in database server, described
Database server adopts existing database server, because described arithmetic server is taken for prior art with database server
Business device, the present invention is to this it will not be described its work process, and the application does not lie in the improvement to algorithm and structuring number yet
According to storage, but the mesh arrangement architecture system of multi-modal data collection is proposed.
As an embodiment, in the present invention, described detection terminal includes one-level detection terminal and secondary detection terminal, institute
State one-level detection terminal and realize the transmission of number data with described data collection station, control centre's communication connection respectively;Described two grades
Detection terminal realizes the transmission of number data with described data collection station, the communication connection of one-level detection terminal respectively.
Wherein, described one-level detection terminal, secondary detection terminal can be realized by control centre by existing known technology
By the transmission situation of the signal of each detection terminal come setup control, can be such as by the detection terminal of energy normal data connection
From being set to one-level detection terminal it is impossible to the detection terminal of normal data connection is from being set to secondary detection terminal;Secondary detection terminal
Carry out signal connection with one-level detection terminal, realize data and transmit to control centre, so can realize detection terminal and control
The transmission of the communication data at center is it is ensured that the stable and reliability of data transfer.
As an embodiment, in the present invention, described control centre include machine learning server and with described machine
The database server that learning server connects, described machine learning server communicates to be connected with detection terminal realizes data
Communication.
The biological attribute data data that described machine learning server is mainly used for detection terminal being formed and storing is led to
Cross existing machine learning techniques and carry out machine learning, the corresponding behavior of study identification people.
Likewise, in the present invention, described machine learning server and the data being connected with described machine learning server
Storehouse server adopts the machine learning server of prior art, database server, former to their work process and work
Reason, because not being the attentinal contents of the application, does not carry out the improvement of internal work, therefore, it will not be described to this for the present invention yet
Its work process, and the application does not lie in the storage of the improvement to algorithm and structural data yet, but proposition one is multi-modal
The mesh arrangement architecture system of data acquisition.
The use control method of the present invention is as follows:
Step one, data collection station are responsible for watching information, acoustic information, coloured image letter attentively by dynamic to depth image and eye
Breath, thermography are as information, and are corrected by the system clock of control centre so that data signal keeps synchronous, with
When will include depth image and eye and move watching information, acoustic information, color image information, thermal imagery information data attentively by communicator
Various types of signal real-time data transmission to detection terminal;
Step 2, detection terminal data acquisition terminal connect, respectively to coloured image, depth image, thermography picture, sound
Message breath and eye move gaze data and are processed, and carry out pattern recognition, shape by the machine learning algorithm of machine learning server
Become to move, based on biological information, Activity recognition information and eye, the structural data watching information attentively, and to these structural datas
Stored, be stored in database server;
Each detection terminal of net distribution transmits signal to control centre simultaneously, and control centre utilizes prior art according to letter
Number transmission situation, the detection terminal that normal data is connected is from being set to one-level detection terminal it is impossible to the detection of normal data connection
Terminal is set to secondary detection terminal certainly;Secondary detection terminal and one-level detection terminal carry out signal connection, realize data to control
Central transmission;
Step 3, the connection situation of control centre's detection zone data, mark which detection terminal can directly and control
Centre data connects, and which detection terminal needs could be with control centre's data communication, and according to inspection by other detection terminals
The framework record surveying terminal determines which fixed test terminal fails and control centre's data communication;
Step 4, such as control centre find there is the detection terminal failing with control centre's data communication, control centre's weight
New planning simultaneously adjusts the setting position of activity detection terminal, and repeat the above steps one are to step 4;As equal in whole detection terminals
With control centre's data communication, stop adjustment work, carry out information transfer data afterwards and be served by.
More than can be seen that, it is single that the present invention solves behavioral value system equipment in prior art, detect simple question,
Achieve depth image and eye moves the stereopsis result watching information attentively, meanwhile, the net distribution using detection terminal solves
The problem of detection information confusing communication is it is achieved that the behavior cognition detection of integrated multi-modal data;Detection terminal can be passed through
Machine learning algorithm carries out pattern recognition, forms more accurate biological information, behavior characteristic information and eye and moves and watch letter attentively
Breath;Present configuration is simple, carries out structural adjustment using existing equipment facility and can achieve, improvement cost is low.
The above is only the preferred embodiment of the present invention it is noted that ordinary skill people for the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (4)
1. a kind of the behavioral value system watching information attentively is moved based on depth image and eye, including control centre it is characterised in that also
Include with described control centre communication connection net distribution multiple detection terminals and with described detection terminal communication link
The data collection station connecing;Described data collection station includes depth image sensor, eye tracker sensor, colour image sensing
Device, sound transducer, thermal imagery sensor.
2. move the behavioral value system watching information attentively according to claim 1 based on depth image and eye it is characterised in that institute
State detection terminal and include arithmetic server and the database server being connected with described arithmetic server.
3. move the behavioral value system watching information attentively according to claim 2 based on depth image and eye it is characterised in that institute
State detection terminal and include one-level detection terminal and secondary detection terminal, described one-level detection terminal is respectively with described data acquisition eventually
The transmission of number data is realized in end, control centre's communication connection;Described secondary detection terminal respectively with described data collection station, one-level
The transmission of number data is realized in detection terminal communication connection.
4. according to any one of claim 1-3, the behavioral value system watching information attentively is moved based on depth image and eye, it is special
Levy and be, described control centre includes machine learning server and the database service being connected with described machine learning server
Device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610837316.5A CN106377265A (en) | 2016-09-21 | 2016-09-21 | Behavior detection system based on depth image and eye movement watching information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610837316.5A CN106377265A (en) | 2016-09-21 | 2016-09-21 | Behavior detection system based on depth image and eye movement watching information |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106377265A true CN106377265A (en) | 2017-02-08 |
Family
ID=57935809
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610837316.5A Withdrawn CN106377265A (en) | 2016-09-21 | 2016-09-21 | Behavior detection system based on depth image and eye movement watching information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106377265A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109620259A (en) * | 2018-12-04 | 2019-04-16 | 北京大学 | Based on eye movement technique and machine learning to the system of autism children's automatic identification |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101753992A (en) * | 2008-12-17 | 2010-06-23 | 深圳市先进智能技术研究所 | Multi-mode intelligent monitoring system and method |
CN101828913A (en) * | 2010-04-16 | 2010-09-15 | 无锡市优特科科技有限公司 | Electrocardiogram remote diagnosis system based on arithmetic server |
CN103198605A (en) * | 2013-03-11 | 2013-07-10 | 成都百威讯科技有限责任公司 | Indoor emergent abnormal event alarm system |
CN104463174A (en) * | 2014-12-16 | 2015-03-25 | 广州南方电力集团科技发展有限公司 | Multi-angle dynamic people recognition and behavior prediction system |
CN105117116A (en) * | 2015-05-06 | 2015-12-02 | 湖南工学院 | System and method for acquiring behavior data generated by executing control task |
CN105787469A (en) * | 2016-03-25 | 2016-07-20 | 广州市浩云安防科技股份有限公司 | Method and system for pedestrian monitoring and behavior recognition |
-
2016
- 2016-09-21 CN CN201610837316.5A patent/CN106377265A/en not_active Withdrawn
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101753992A (en) * | 2008-12-17 | 2010-06-23 | 深圳市先进智能技术研究所 | Multi-mode intelligent monitoring system and method |
CN101828913A (en) * | 2010-04-16 | 2010-09-15 | 无锡市优特科科技有限公司 | Electrocardiogram remote diagnosis system based on arithmetic server |
CN103198605A (en) * | 2013-03-11 | 2013-07-10 | 成都百威讯科技有限责任公司 | Indoor emergent abnormal event alarm system |
CN104463174A (en) * | 2014-12-16 | 2015-03-25 | 广州南方电力集团科技发展有限公司 | Multi-angle dynamic people recognition and behavior prediction system |
CN105117116A (en) * | 2015-05-06 | 2015-12-02 | 湖南工学院 | System and method for acquiring behavior data generated by executing control task |
CN105787469A (en) * | 2016-03-25 | 2016-07-20 | 广州市浩云安防科技股份有限公司 | Method and system for pedestrian monitoring and behavior recognition |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109620259A (en) * | 2018-12-04 | 2019-04-16 | 北京大学 | Based on eye movement technique and machine learning to the system of autism children's automatic identification |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107679522B (en) | Multi-stream LSTM-based action identification method | |
CN104966327A (en) | System and method for monitoring health and registering attendance on basis of internet of things | |
CN110135476A (en) | A kind of detection method of personal safety equipment, device, equipment and system | |
CN109241917A (en) | A kind of classroom behavior detection system based on computer vision | |
US20200254299A1 (en) | System and non-intrusive method for exercise posture detection | |
CN105554079A (en) | Remote manual product quality detection method based on mobile terminals | |
CN110049121A (en) | A kind of data center's cruising inspection system based on augmented reality | |
CN106125909A (en) | A kind of motion capture system for training | |
CN108804971A (en) | A kind of image identification system, augmented reality show equipment and image-recognizing method | |
CN204465592U (en) | A kind of campus monitoring management system based on Internet of Things | |
CN108764640A (en) | A kind of personalized dynamic teaching diagnosis and the system of improvement based on big data | |
CN107566790A (en) | With reference to REID and the real time monitoring apparatus and method of Video Supervision Technique | |
CN108597264A (en) | A kind of tutoring system and its implementation tracing review | |
CN108769442A (en) | A kind of hospital's calling system | |
CN110345407A (en) | A kind of intelligent posture correction desk lamp and strong attitude positioning method based on deep learning | |
CN109766790A (en) | A kind of pedestrian detection method based on self-adaptive features channel | |
CN107705658A (en) | A kind of new education system | |
CN107610260B (en) | Intelligent attendance system and attendance method based on machine vision | |
CN106377265A (en) | Behavior detection system based on depth image and eye movement watching information | |
CN107330822A (en) | A kind of children's safety KXG and method | |
Zhou | Construction of new intelligent teaching scene based on multimodal wireless sensing technology | |
CN108109464A (en) | A kind of shared bicycle learning system based on VR technologies | |
CN206365889U (en) | Human body based on BLE Mesh is sat, stance correction system | |
KR102234082B1 (en) | Hybrid IoT Management type desk for reading room, and Reading room learning management system using the same | |
CN106525023A (en) | Array localization device and array localization method based on data analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170208 |