CN109543517A - A kind of computer vision artificial intelligence application method and system - Google Patents
A kind of computer vision artificial intelligence application method and system Download PDFInfo
- Publication number
- CN109543517A CN109543517A CN201811203698.1A CN201811203698A CN109543517A CN 109543517 A CN109543517 A CN 109543517A CN 201811203698 A CN201811203698 A CN 201811203698A CN 109543517 A CN109543517 A CN 109543517A
- Authority
- CN
- China
- Prior art keywords
- trunk
- image
- artificial intelligence
- computer vision
- inclined degree
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
- G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0438—Sensor means for detecting
- G08B21/0476—Cameras to detect unsafe condition, e.g. video cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Human Computer Interaction (AREA)
- Psychology (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of computer vision artificial intelligence application methods, comprising: fitting depth camera is in the region of required monitoring；By system interface activation system, system connects kinect automatically；By the trunk inclined degree of depth camera real-time detection all human bodies within the vision, bone image is obtained；According to the bone image of acquisition, trunk and ground angulation are calculated；According to trunk and ground angulation, it is compared with preset threshold；Under the three-dimensional coordinate to specified directory for automatically saving the depth map at corresponding moment, skeletal graph and all skeleton points, when trunk and ground angulation are more than threshold value, early warning is issued.The present invention innovatively passes through monitoring trunk inclined degree and carries out falling down early warning, can be realized the purpose for reducing and falling down generation.
Description
Technical field
The present invention relates to artificial intelligence field more particularly to a kind of computer vision artificial intelligence application method and system.
Background technique
Computer vision refers to replaces human eye the calculating such as to be identified, tracked and measure to target using video camera and computer
Machine vision, and graphics process is further done, so that computer is treated as the image for being more suitable for eye-observation or sending instrument detection to.
Computer vision mainly distinguishes four steps: image acquisition, image rectification, Stereo matching and three-dimensional reconstruction.Under normal circumstances, people
Class obtains image by eyes, eyes can be approximately it is arranged in parallel, when observing Same Scene, left eye obtains the scene on the left side
Information is more, to the right in left view nethike embrane, and the scene information that right eye obtains the right is more, to the left in right retina.It is same
Image point locations difference of the scene point on left view nethike embrane and on right retina is parallax, is the important letter for perceiving Object Depth
Breath.The principle that computer vision obtains image is similar to the mankind, is that different images is obtained by the camera on different location,
The image of left video camera shooting is known as left image, and the image of right video camera shooting is known as right image.Left image obtains the field on the left side
Scape information is more, and the scene information that right image obtains the right is more.In image acquisition procedures, it will lead to there are many factor
Image fault, as the aberration of imaging system, distortion, bandwidth it is limited etc. caused by image fault；Since image device shoots posture
With image geometric distortion caused by scan non-linearity；The image as caused by motion blur, radiation distortion, introducing noise etc. loses
Very.In the image of Same Scene shoot under two width or several different locations and corresponding, relationship between Matching unit is established
Process is known as Stereo matching.Such as in binocular solid matching, Matching unit selects pixel, then obtains and corresponds to Same Scene
Two images in matched pixel position difference, i.e. parallax.Parallax is transformed between 0-255 in proportion, with grayscale image
Form shows, as disparity map.According to the parallax for the pixel that Stereo matching obtains, if it is known that the inside and outside ginseng of camera
Number, then obtain the depth information of object in scene according to video camera geometrical relationship, and then obtain the three-dimensional coordinate of object in scene.
Artificial intelligence is the research movable rule of human intelligence, constructs the manual system with certain intelligence, how is research
Allow computer to go the intelligence for completing to need people in the past just competent work, that is, study how the software and hardware of appliance computer
To simulate basic theories, the methods and techniques of the certain behaviors of the mankind.The principle of artificial intelligence are as follows: computer can by sensor or
The mode being manually entered come the fact that collect about some scene, this information is compared by computer with stored information,
With its meaning of determination.Computer can calculate various possible movements according to the information being collected into, and predict the effect of which kind of movement
It is best.
The tracking of Kinect bone uses TOF technology.Infrared transmitter actively projects modulated near infrared light, infrared light
Fibre is found will reflect on the object in the visual field, and infrared camera receives reflected infrared ray, be measured using TOF technology
Depth calculates the time difference (usually being calculated by phase difference) of light, according to depth (the i.e. object to depth that can obtain object
The distance of camera).The 3D depth image that Microsoft will detect, is transformed into skeleton tracing system.Kinect is found in image compared with can
It can be the object of human body, then be assessed depth image (machine learning) different parts to differentiate human body.Process flow
Final step be using last stage export as a result, generating a width shell system according to 20 artis tracked.
Each possible pixel that Kinect can assess Exemplar output determines artis, in this way, Kinect energy
It is enough that the actually located position of human body is most accurately assessed based on sufficient information.In addition the Model Matching stage has also done some additional
Output filter smoothly exports and handles the special events such as occlusion joint.
In the prior art, it generally detects whether to fall down by detection trunk falling head and speed, lack logical
Cross correlation technique and system that monitoring data carry out early warning tumble.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of computer vision artificial intelligence application sides
Method.The present invention is based on attitude detections to be judged simultaneously for disadvantaged group such as old men by detecting the inclined degree of its trunk
Provide early warning, the generation of fall prevention event.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of computer vision artificial intelligence application method, specific steps include:
Fitting depth camera is in the region of required monitoring；By system interface activation system, system connects automatically
kinect；
By the trunk inclined degree of depth camera real-time detection all human bodies within the vision, skeletal graph is obtained
Picture；
According to the bone image of acquisition, trunk and ground angulation are calculated；
According to trunk and ground angulation, it is compared with preset threshold；
Under the three-dimensional coordinate to specified directory for automatically saving the depth map at corresponding moment, skeletal graph and all skeleton points, when
Trunk and ground angulation are more than threshold value, issue early warning.
Specifically, the present invention monitors all people's body number no more than 6 within sweep of the eye.
Specifically, the depth map of preservation is bmp format, and the skeletal graph of preservation is bmp format, and the skeleton point three-dimensional of preservation is sat
It is designated as txt format.
It is possible to further press from both sides button by the select file in interface come the specified required catalogue saved.
Further, include a text display box on the right side of interface, be used for real-time display trunk inclined degree.
Another object of the present invention is to provide a kind of computer vision artificial intelligence application systems.
Another object of the present invention can be achieved through the following technical solutions:
A kind of computer vision artificial intelligence application system, including data processing module, data acquisition module, display module
And warning module.
The data acquisition module is kinect depth camera, real for monitoring all human bodies within sweep of the eye automatically
When monitor its trunk inclined degree；
The data processing module, for handling the trunk inclined degree data of acquisition, according to data system
Make the three-dimensional coordinate of the depth map at corresponding moment, skeletal graph and all skeleton points；
The display module includes system interface, for by interface come activation system；The image display box on the interface left side
For real-time display depth image；Image display box on the right of interface is used for real-time display bone image；
The warning module, for providing early warning automatically when processing module detects that inclined degree is more than a certain threshold value,
Or related personnel is notified to handle after falling down.
The present invention compared to the prior art, have it is below the utility model has the advantages that
The present invention is for the hazardous act in life and work, such as bends over, takes that be higher by object trunk obliquity effects body flat
Weighing apparatus causes to be easy to happen the behavior fallen down, and carries out detection and early warning, reduces the generation fallen down.Even after falling down, this system
Also related personnel can be notified to handle in time.The present invention innovatively passes through monitoring trunk inclined degree and carries out falling down early warning,
It can be realized the purpose for reducing and falling down generation.
Detailed description of the invention
Fig. 1 is schematic structural diagram of the device of the invention；
Fig. 2 is the flow chart of the method for the present invention.
Specific embodiment
Present invention will now be described in further detail with reference to the embodiments and the accompanying drawings, but embodiments of the present invention are unlimited
In this.
Embodiment
It is as shown in Figure 1 a kind of flow chart of computer vision artificial intelligence application method, specific steps include:
(1) fitting depth camera is in the region of required monitoring；By system interface activation system, system connects automatically
kinect；
(2) the trunk inclined degree for passing through depth camera real-time detection all human bodies within the vision, obtains bone
Image；
(3) according to the bone image of acquisition, trunk and ground angulation are calculated；
(4) it according to trunk and ground angulation, is compared with preset threshold；
(5) under the three-dimensional coordinate to specified directory for automatically saving the depth map at corresponding moment, skeletal graph and all skeleton points,
When trunk and ground angulation are more than threshold value, sending early warning.
Specifically, the present invention monitors all people's body number no more than 6 within sweep of the eye.
Specifically, the depth map of preservation is bmp format, and the skeletal graph of preservation is bmp format, and the skeleton point three-dimensional of preservation is sat
It is designated as txt format.
It is possible to further press from both sides button by the select file in interface come the specified required catalogue saved.
Further, include a text display box on the right side of interface, be used for real-time display trunk inclined degree.
It is illustrated in figure 2 a kind of structure chart of computer vision artificial intelligence application system, the system comprises at data
Manage module, data acquisition module, display module and warning module.
The data acquisition module is kinect depth camera, real for monitoring all human bodies within sweep of the eye automatically
When monitor its trunk inclined degree；
The data processing module, for handling the trunk inclined degree data of acquisition, according to data system
Make the three-dimensional coordinate of the depth map at corresponding moment, skeletal graph and all skeleton points；
The display module includes system interface, for by interface come activation system；The image display box on the interface left side
For real-time display depth image；Image display box on the right of interface is used for real-time display bone image；
The warning module, for providing early warning automatically when processing module detects that inclined degree is more than a certain threshold value,
Or related personnel is notified to handle after falling down.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, other any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present invention,
It should be equivalent substitute mode, be included within the scope of the present invention.
Claims (4)
1. a kind of computer vision artificial intelligence application method, which is characterized in that specific steps include:
Fitting depth camera is in the region of required monitoring；By system interface activation system, system connects kinect automatically；
By the trunk inclined degree of depth camera real-time detection all human bodies within the vision, bone image is obtained；
According to the bone image of acquisition, trunk and ground angulation are calculated；
According to trunk and ground angulation, it is compared with preset threshold；
Under the three-dimensional coordinate to specified directory for automatically saving the depth map at corresponding moment, skeletal graph and all skeleton points, work as trunk
It is more than threshold value with ground angulation, issues early warning.
2. a kind of computer vision artificial intelligence application method according to claim 1, which is characterized in that in the method
The depth map of preservation is bmp format, and the skeletal graph of preservation is bmp format, and the skeleton point three-dimensional coordinate of preservation is txt format.
3. a kind of computer vision artificial intelligence application method according to claim 1, which is characterized in that the automatic guarantor
Depth map, skeletal graph and three coordinates deposited can press from both sides button by the select file in interface come specified required preservation
Catalogue.
4. a kind of computer vision artificial intelligence application system for realizing claim 1-3, which is characterized in that the system
Including data processing module, data acquisition module, display module and warning module；
The data acquisition module is kinect depth camera, for monitoring all human bodies within sweep of the eye automatically, is supervised in real time
Survey its trunk inclined degree；
The data processing module, for handling the trunk inclined degree data of acquisition, according to data creating phase
Answer the three-dimensional coordinate of the depth map at moment, skeletal graph and all skeleton points；
The display module includes system interface, for by interface come activation system；The image display box on the interface left side is used for
Real-time display depth image；Image display box on the right of interface is used for real-time display bone image；
The warning module, for when processing module detect inclined degree be more than a certain threshold value when, provide early warning automatically, or
Related personnel is notified to handle after falling down.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811203698.1A CN109543517A (en) | 2018-10-16 | 2018-10-16 | A kind of computer vision artificial intelligence application method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811203698.1A CN109543517A (en) | 2018-10-16 | 2018-10-16 | A kind of computer vision artificial intelligence application method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109543517A true CN109543517A (en) | 2019-03-29 |
Family
ID=65843926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811203698.1A Pending CN109543517A (en) | 2018-10-16 | 2018-10-16 | A kind of computer vision artificial intelligence application method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109543517A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111104932A (en) * | 2020-02-03 | 2020-05-05 | 北京都是科技有限公司 | Tumble detection system and method and image processor |
| CN114419842A (en) * | 2021-12-31 | 2022-04-29 | 浙江大学台州研究院 | Artificial intelligence-based falling alarm method and device for assisting user in moving to intelligent closestool |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1874497A (en) * | 2006-05-30 | 2006-12-06 | 浙江工业大学 | Household safe and security equipment for solitary old person based on omnibearing computer vision |
| CN105335696A (en) * | 2015-08-26 | 2016-02-17 | 湖南信息职业技术学院 | 3D abnormal gait behavior detection and identification based intelligent elderly assistance robot and realization method |
| CN105719429A (en) * | 2014-07-29 | 2016-06-29 | 吴诗蕊 | Fall detection and alarm system based on Kinect and operating method thereof |
| CN107194967A (en) * | 2017-06-09 | 2017-09-22 | 南昌大学 | Human fall detection method and device based on Kinect depth image |
-
2018
- 2018-10-16 CN CN201811203698.1A patent/CN109543517A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1874497A (en) * | 2006-05-30 | 2006-12-06 | 浙江工业大学 | Household safe and security equipment for solitary old person based on omnibearing computer vision |
| CN105719429A (en) * | 2014-07-29 | 2016-06-29 | 吴诗蕊 | Fall detection and alarm system based on Kinect and operating method thereof |
| CN105335696A (en) * | 2015-08-26 | 2016-02-17 | 湖南信息职业技术学院 | 3D abnormal gait behavior detection and identification based intelligent elderly assistance robot and realization method |
| CN107194967A (en) * | 2017-06-09 | 2017-09-22 | 南昌大学 | Human fall detection method and device based on Kinect depth image |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111104932A (en) * | 2020-02-03 | 2020-05-05 | 北京都是科技有限公司 | Tumble detection system and method and image processor |
| CN114419842A (en) * | 2021-12-31 | 2022-04-29 | 浙江大学台州研究院 | Artificial intelligence-based falling alarm method and device for assisting user in moving to intelligent closestool |
| CN114419842B (en) * | 2021-12-31 | 2024-05-10 | 浙江大学台州研究院 | Fall alarm method and device for assisting user to fall to closestool based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6690041B2 (en) | Method and device for determining point of gaze on three-dimensional object | |
| US11030455B2 (en) | Pose recognition method, device and system for an object of interest to human eyes | |
| CN107004275B (en) | Method and system for determining spatial coordinates of a 3D reconstruction of at least a part of a physical object | |
| CN106843507B (en) | Virtual reality multi-person interaction method and system | |
| CA2620619C (en) | Capturing and processing facial motion data | |
| EP2064676B1 (en) | A method and system for three-dimensional model acquisition | |
| CN103099602B (en) | Based on the physical examinations method and system of optical identification | |
| Duchowski et al. | 3-D eye movement analysis | |
| KR20150117553A (en) | Method, apparatus and computer readable recording medium for eye gaze tracking | |
| KR101471488B1 (en) | Device and Method for Tracking Sight Line | |
| CN113808160B (en) | Sight direction tracking method and device | |
| CN110598590A (en) | Close interaction human body posture estimation method and device based on multi-view camera | |
| Munn et al. | 3D point-of-regard, position and head orientation from a portable monocular video-based eye tracker | |
| EP1627294A2 (en) | A man-machine interface based on 3-d positions of the human body | |
| CN107861625A (en) | Gaze tracking system and method based on 3d space model | |
| CN109035307B (en) | Set area target tracking method and system based on natural light binocular vision | |
| CN106840112A (en) | A kind of space geometry measuring method of utilization free space eye gaze point measurement | |
| CN102932638B (en) | 3D video monitoring method based on computer modeling | |
| CN101581874A (en) | Tele-immersion teamwork device based on multi-camera acquisition | |
| CN108205314A (en) | Based on the matched robot navigation device of stereoscopic vision and system | |
| CN108881885A (en) | Advanced treatment system | |
| CN109213138A (en) | A kind of barrier-avoiding method, apparatus and system | |
| CN109543517A (en) | A kind of computer vision artificial intelligence application method and system | |
| Štrbac et al. | Kinect in neurorehabilitation: computer vision system for real time hand and object detection and distance estimation | |
| CN111489392A (en) | Single target human motion posture capturing method and system in multi-person environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190329 |