CN105975956A - Infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone - Google Patents
Infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone Download PDFInfo
- Publication number
- CN105975956A CN105975956A CN201610367782.1A CN201610367782A CN105975956A CN 105975956 A CN105975956 A CN 105975956A CN 201610367782 A CN201610367782 A CN 201610367782A CN 105975956 A CN105975956 A CN 105975956A
- Authority
- CN
- China
- Prior art keywords
- human body
- target
- infrared
- track
- point
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/35—Categorising the entire scene, e.g. birthday party or wedding scene
- G06V20/36—Indoor scenes
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Abstract
The invention, which belongs to the technical field of the video image, discloses an infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone. The method comprises: step one, an infrared panorama pick-up head is used for carrying out infrared shooting, thereby obtaining a video image signal; step two, an infrared panoramic picture is unfolded by using a fast approximate unfolding method; step three, an indoor grid semantic map in a home environment is constructed; step four, on the basis of an improved hybrid Gaussian model algorithm, modeling of a human body target is carried out, a target block mass is identifier, block mass information is obtained, the target block mass is tracked in real time, and a human body track feature is obtained; step five, a human body moving track is extracted by using a hybrid Gaussian clustering method; step six, the human body track is segmented; and step seven, a convolutional neural network is trained by using the track after human body motion segmentation in the home environment as a training sample, feature extraction is carried out on a human body behavior track in the home environment, and classification is carried out by using evidence reasoning; and if an abnormal situation occurs, alarming is carried out.
Description
Technical field
The invention belongs to video image technical field, relate to a kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head.
Background technology
Fast development along with economic society, living standards of the people and the huge improvement of Medical health guarantee cause, fertility rate persistently keeps reduced levels, the situation is tense in Chinese society aging, 2013, and more than the 60 years old aged of China is more than 200,000,000, within 2025,300,000,000 will be broken through, the year two thousand forty will break through 400,000,000, and within more than 80 years old when the time comes, old man is up to 1 hundred million, and aging will affect economy and the society of China for a long time.Due to the enforcement of family planning policy, the traditional nucleus family's structure of China changes, and household size small has broken the home mode that three generations people traditionally even four generation people lives together.In modern society, old man and children are desirable that " free space " of oneself simultaneously, and therefore the social phenomenon of China's Empty nest elderly and old solitary people is the most prominent.China's empty-nested elderly people mouth in 2012 is 0.99 hundred million people according to statistics, within 2013, breaks through 100,000,000 high pointes.With advancing age, also along with ergasia reduced capability while the organism physiology of old people is aging, so that its life produces a lot of potential safety hazard, falling down easily occurs, the unexpected injury such as falls.The unexpected common space in a newspaper of report that Nobody Knows of dying many days of being even in is there is in old solitary people because of unmanned nurse.Therefore, how to detect Deviant Behavioies such as falling down under the home environment of old solitary people in time, the very first time notifies household and hospital, ensures the safety of old solitary people, is the important topic being worth us to pay close attention to.
At present, unusual checking is a study hotspot, has the biggest development space.In the prior art, there are a kind of human body unusual checking based on action recognition and early warning system, this application scheme proposes by camera collection human motion image, Inertial Measurement Unit in human body muffetee gathers human body movement data, and it is respectively sent to main frame, combine human body movement data by main frame plug-in according to human motion image and simulate relevant action picture over the display, when there being Deviant Behavior to occur, can be reported to the police by early warning system.Moving image is combined by this technology with exercise data, and to realize the detection of human body Deviant Behavior, but the muffetee that human hands is worn belongs to contact device, when gathering data, the normal activity of human body can cause certain impact so that the practicality of this technology is relatively low.Meanwhile, also class behavior monitoring based on pure digi-tal image or video information recognition methods, the pixel characteristic change that normal behaviour and Deviant Behavior are brought from image finds and extract rule, thus realizes abnormality and find.As: a kind of by processing the method that the gait information in image identifies the state of people, the method includes parametric method and imparametrization method, parametric method distinguishes people based on the cadence in gait Time And Space Parameters and step-length, and imparametrization method feature based face technology identification eigengait distinguishes people.But, existing various scheme many employings common camera directly monitor, and be difficult to meet the demand of Human bodys' response, and be difficult to protect the privacy of monitored object in terms of areas imaging, resolution.The limited visual range of common camera limits the acquisition of information, if visual range can be expanded, will solve the problem of acquisition of information easily.Infrared panorama photographic head can realize 360 ° of all standings to monitoring environment, and visual range is wide, avoids the impact of the factor such as indoor illumination, shade simultaneously.Therefore, set forth herein that a kind of visual range is wide, untouchable, to have certain secret protection effect Deviant Behavior monitoring method is monitored to the Deviant Behavior realizing old solitary people under home environment.
Summary of the invention
In view of this, it is an object of the invention to provide a kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head, the method is based on infrared panorama camera technique, Blob image tracking algorithm, convolutional neural networks and DS evidential reasoning sorting algorithm, the daily behavior range of activity of old solitary people can be realized all standing formula tracing detection, and carry out alarm when causing danger abnormal conditions.
For reaching above-mentioned purpose, the present invention provides following technical scheme:
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head, belongs to video image technical field.Comprise the following steps: step one: utilize panorama infrared camera to carry out infrared photography, obtain video signal；Step 2: use quick approximate expansion method to infrared panorama image spread；Step 3: build indoor grille semanteme map under home environment；Step 4: use and model human body target based on the mixed Gauss model algorithm improved, identifies targeted mass, obtains agglomerate information, and targeted mass carries out real-time tracking, obtains human body track characteristic；Step 5: use mixed Gaussian clustering method to extract human body motion track；Step 6: human body track is split；Step 7: using the track after human body motion segmentation under home environment as training sample training convolutional neural networks, human body action trail under home environment being carried out feature extraction, evidential reasoning is classified；Report to the police if abnormal.
Further, described step 2 use quick approximate expansion method to infrared panorama image spread.
Further, choosing general adult normal step-length 50cm in step 3 is a grid base our unit, set up indoor coordinate system, target place to be identified indoor environment is set up grid semanteme map, wherein map includes each room indoor and indoor main furniture, the title of equipment, coordinate, furniture and equipment in room are the key point of human body motion track.
Further, step 4 specifically includes following steps: 41: set up mixture Gaussian background model and characterize the feature of each pixel in picture frame；42: adaptive updates background model, calculate foreground mask；43: foreground mask is carried out morphology opening operation and closed operation filtering；44: use the human body target in agglomerate detection algorithm based on foreground mask connected region detection video area, obtain agglomerate information, i.e. human body target information；45: use MeanShift algorithm that human body target is carried out real-time tracking.
Further, described step 5 use mixed Gaussian clustering method extract human body track.
Further, following steps are specifically included: after obtaining human body motion track in step 6, first determine whether the key point belonging to currently putting, calculate each topology point and belong to the probability of this key point, select the topological point that maximum of probability is corresponding, if this maximum of probability is more than a certain threshold value, if then thinking that current point belongs to topology point. current point belongs to topology point, then this point of labelling is the beginning or end of sequence, continual behavior sequence is divided into the behavior sequence with key point as beginning and end, and wherein track characteristic also includes that target is at the treated duration of terminal.
Further, step 7 specifically includes following steps: 71: using the behavior sequence of different target that obtained as training sample, convolutional neural networks is carried out pre-training；72: with the daily routines action trail of target to be identified, convolutional neural networks model is finely adjusted, it is thus achieved that concrete action trail analyzes model；73: analyze model with the action trail of this target and the action trail obtained tentatively is identified, construct target recognition rate matrix；74: using DS evidential reasoning to merge composition rule and human body behavior is carried out Classification and Identification, if the action trail terminal of target is in non-critical areas, and duration exceedes threshold value；Or the starting and terminal point of target round number of times in certain time length exceedes threshold value and is Deviant Behavior, now send urgent message to household and the hospital of old man.
Further, the agglomerate information described in step 4 includes: the center of agglomerate, area and id information.
Further, sending urgent message described in step 5 is that SMS module based on the Internet realizes.
The beneficial effects of the present invention is: the method for the invention is based on infrared photography skill; Blob image tracking algorithm; convolutional neural networks and DS evidential reasoning sorting algorithm; wherein; infrared photography function detects thermal target exactly; and ignore the detailed information that background and human body itself are abundant, therefore the privacy of monitored object can be played certain protective effect；Full-view camera is capable of 360 ° of all standings in the range of physical activity, and visual range is wide, avoids the impact of the factor such as indoor illumination, shade simultaneously, obtains more movable information and details；In track algorithm based on Blob, the information of targeted mass can be obtained the most easily so that real-time track algorithm is achieved；Feature extraction algorithm based on convolutional neural networks passes through degree of depth network extraction high-level semantics features, is obtained from meeting market's demand by unsupervised learning, simultaneously, stability is high, grey scale change to image, influence of noise and viewpoint change are insensitive, have the Invariance feature such as translation, rotation.Deviant Behavior criterion is simply effectively reliable so that the method can realize falling the purpose of unusual checking.Meanwhile, utilize infrared panorama video camera to carry out behavior monitoring, it is achieved that by untouchable mode, carry out the purpose of Deviant Behavior monitoring comprehensively, and the track algorithm real-time that the present invention uses is good, follow the tracks of target characteristic and be easily obtained, it is judged that standard is simple and reliable.
Accompanying drawing explanation
In order to make the purpose of the present invention, technical scheme and beneficial effect clearer, the present invention provides drawings described below to illustrate:
Fig. 1 is the flow chart of the method for the invention；
Fig. 2 is indoor grille of the present invention figure schematic diagram semantically；
Fig. 3 is convolutional neural networks identification process figure of the present invention.
Detailed description of the invention
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.
Fig. 1 is the flow chart of the method for the invention, and this method comprises the following steps:
S1: utilize panorama infrared camera to carry out infrared pick-up, obtain video signal.The installation of video camera, according to real space layout distribution situation, is arranged in the ceiling central authorities in each room, accomplishes the comprehensive covering to this room and target living environment to be monitored.
S2: the panoramic picture obtaining 360 ° of photographic head of infrared panorama carries out reduction and launches, and solves the imaging distortion problem in panoramic picture；Use quick approximate expansion method to infrared panorama image spread in the present embodiment.
S3: choosing general adult normal step-length 50cm is a grid base our unit, set up indoor coordinate system, target place to be identified indoor environment is set up grid semanteme map, wherein map includes each room indoor and indoor main furniture, the title of equipment, coordinate, furniture and equipment in room are the key point of human body motion track.
S4: use ADAPTIVE MIXED Gauss modeling method to carry out foreground detection, the ADAPTIVE MIXED Gauss modeling method in described step S2, specifically comprise the following steps that
S41: set up mixture Gaussian background model and characterize the feature of each pixel in picture frame；
S42: adaptive updates background model, calculates foreground mask；
S43: foreground mask is carried out morphology opening operation and closed operation filtering.
S44: identify the human body target in region to be detected, obtains human body target information (center, area, ID)；Use agglomerate detection algorithm identification human body target based on foreground mask connected region in the present embodiment.
S45: targeted mass is carried out real-time tracking；Use MeanShift algorithm that targeted mass is carried out real-time tracking in the present embodiment.
S5: the behavior to human body target is estimated to split, and track sets is converted into the key point sequence represented by key point；Mixed Gaussian clustering method is used to extract human motion key point sequence in the present embodiment.
S6: after obtaining human body motion track, first determine whether the key point belonging to currently putting, calculate each topology point and belong to the probability of this key point, select the topological point that maximum of probability is corresponding, if this maximum of probability is more than a certain threshold value, if then thinking that current point belongs to topology point. current point belongs to topology point, then this point of labelling is the beginning or end of sequence, continual behavior sequence is divided into the behavior sequence with key point as beginning and end, wherein track characteristic also includes that target is at the treated duration of terminal, the most round number of times between Origin And Destination.
S7: use convolutional neural networks that human body target carries out trajectory analysis, and with DS Evidential reasoning algorithm, target behavior is carried out Classification and Identification.If the action trail terminal of target is in non-critical areas, and duration exceedes threshold value；Or the starting and terminal point of target round number of times in certain time length exceedes threshold value and is Deviant Behavior, now send urgent message to household and the hospital of old man；The concrete step identified is as follows:
S71: using the behavior sequence of different target that obtained as training sample, convolutional neural networks is carried out pre-training；
S72: convolutional neural networks model is finely adjusted with the daily routines action trail of target to be identified, it is thus achieved that concrete action trail analyzes model；
S73: analyze model with the action trail of this target and the action trail obtained tentatively is identified, construct target recognition rate matrix；
S74: using DS evidential reasoning to merge composition rule and human body behavior is carried out Classification and Identification, if the action trail terminal of target is in non-critical areas, and duration exceedes threshold value；Or the starting and terminal point of target round number of times in certain time length exceedes threshold value and is Deviant Behavior, now send urgent message to household and the hospital of old man.
Finally illustrate is, preferred embodiment above is only in order to illustrate technical scheme and unrestricted, although the present invention being described in detail by above preferred embodiment, but skilled artisan would appreciate that, in the form and details it can be made various change, without departing from claims of the present invention limited range.
Claims (9)
1. an old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head, it is characterised in that: comprise the following steps:
Step one: utilize panorama infrared camera to carry out infrared pick-up, obtain video signal；
Step 2: use quick approximate expansion method to infrared panorama image spread；
Step 3: choosing general adult normal step-length 50cm is a grid base our unit, set up indoor coordinate system, target place to be identified indoor environment is set up grid semanteme map, and wherein map includes each room indoor and indoor main furniture, the title of equipment, coordinate；
Step 4: use the ADAPTIVE MIXED Gauss modeling method improved to carry out foreground detection, identify targeted mass, obtain human body target information (center, area and id information), use MeanShift algorithm to carry out real-time tracking for entering the human body target of image pickup scope；
Step 5: use mixed Gaussian clustering method to extract human body motion track；
Step 6: human body track is split；
Step 7: by the track vector after human body motion segmentation under home environment, as training sample training convolutional neural networks, carries out feature extraction to human body action trail under home environment, uses DS evidential reasoning to classify；If the action trail terminal of target is in non-critical areas, and duration exceedes threshold value；Or the starting and terminal point of target round number of times in certain time length exceedes threshold value and is Deviant Behavior, now send urgent message to household and the hospital of old man.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterised in that: described step 2 use quick approximate expansion method to infrared panorama image spread.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterized in that: choosing general adult normal step-length 50cm in described step 3 is a grid base our unit, set up indoor coordinate system, target place to be identified indoor environment is set up grid semanteme map, wherein map includes each room indoor and indoor main furniture, the title of equipment, coordinate.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterised in that: in step 4, specifically include following steps: 41: set up mixture Gaussian background model and characterize the feature of each pixel in picture frame；42: adaptive updates background model, calculate foreground mask；43: foreground mask is carried out morphology opening operation and closed operation filtering；44: use the human body target in agglomerate detection algorithm based on foreground mask connected region detection video area, obtain agglomerate information, i.e. human body target information；45: use MeanShift algorithm that human body target is carried out real-time tracking.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterised in that: in step 5, human body motion track is converted to human motion key point sequence by mixed Gaussian clustering method.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterized in that: step 6 specifically includes following steps: after obtaining human body motion track, first determine whether the key point belonging to currently putting, calculate each topology point and belong to the probability of this key point, select the topological point that maximum of probability is corresponding, if this maximum of probability is more than a certain threshold value, if then thinking that current point belongs to topology point. current point belongs to topology point, then this point of labelling is the beginning or end of sequence, continual behavior sequence is divided into the behavior sequence with key point as beginning and end, wherein track characteristic also includes that target is at terminal treated duration T.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterized in that: in step 7, specifically include following steps: 71: using the behavior sequence of different target that obtained as training sample, convolutional neural networks is carried out pre-training；72: with the daily routines action trail of target to be identified, convolutional neural networks model is finely adjusted, it is thus achieved that concrete action trail analyzes model；73: analyze model with the action trail of this target and the action trail obtained tentatively is identified, construct target recognition rate matrix；74: using DS evidential reasoning to merge composition rule and human body behavior is carried out Classification and Identification, if the action trail terminal of target is in non-critical areas, and duration exceedes threshold value；Or the starting and terminal point of target round number of times in certain time length exceedes threshold value and is Deviant Behavior, now send urgent message to household and the hospital of old man.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterised in that: the agglomerate information described in step 4 includes: the center of agglomerate, area and id information.
A kind of old solitary people Deviant Behavior recognition methods based on infrared panorama photographic head the most according to claim 1, it is characterised in that: sending urgent message in described step 7 is that SMS module based on the Internet realizes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610367782.1A CN105975956A (en) | 2016-05-30 | 2016-05-30 | Infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610367782.1A CN105975956A (en) | 2016-05-30 | 2016-05-30 | Infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105975956A true CN105975956A (en) | 2016-09-28 |
Family
ID=56956204
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610367782.1A Pending CN105975956A (en) | 2016-05-30 | 2016-05-30 | Infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105975956A (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106568445A (en) * | 2016-11-11 | 2017-04-19 | 吉林大学 | Indoor track prediction method based on bidirectional circulation neural network |
| CN106571014A (en) * | 2016-10-24 | 2017-04-19 | 上海伟赛智能科技有限公司 | Method for identifying abnormal motion in video and system thereof |
| CN106780566A (en) * | 2016-11-11 | 2017-05-31 | 华南智能机器人创新研究院 | A kind of method and system of the target following under the robot based on Delta |
| CN106846729A (en) * | 2017-01-12 | 2017-06-13 | 山东大学 | A kind of fall detection method and system based on convolutional neural networks |
| CN106971497A (en) * | 2017-05-03 | 2017-07-21 | 成都宏软科技实业有限公司 | A kind of intelligent safety and defence system based on intelligent domestic |
| CN107063258A (en) * | 2017-03-07 | 2017-08-18 | 重庆邮电大学 | A kind of mobile robot indoor navigation method based on semantic information |
| CN107145878A (en) * | 2017-06-01 | 2017-09-08 | 重庆邮电大学 | Old man's anomaly detection method based on deep learning |
| CN107153871A (en) * | 2017-05-09 | 2017-09-12 | 浙江农林大学 | Fall detection method based on convolutional neural networks and mobile phone sensor data |
| CN107705334A (en) * | 2017-08-25 | 2018-02-16 | 北京图森未来科技有限公司 | A kind of video camera method for detecting abnormality and device |
| CN107766852A (en) * | 2017-12-06 | 2018-03-06 | 电子科技大学 | A kind of man-machine mouse track detection method based on convolutional neural networks |
| CN108229440A (en) * | 2018-02-06 | 2018-06-29 | 北京奥开信息科技有限公司 | One kind is based on Multi-sensor Fusion indoor human body gesture recognition method |
| CN108446594A (en) * | 2018-02-11 | 2018-08-24 | 四川省北青数据技术有限公司 | Emergency reaction ability assessment method based on action recognition |
| CN108537105A (en) * | 2018-02-05 | 2018-09-14 | 中国农业大学 | A kind of hazardous act recognition methods under home environment |
| CN109002858A (en) * | 2018-07-23 | 2018-12-14 | 合肥工业大学 | A kind of clustering ensemble method based on evidential reasoning for user behavior analysis |
| CN109308444A (en) * | 2018-07-16 | 2019-02-05 | 重庆大学 | A kind of abnormal behaviour recognition methods under indoor environment |
| CN109447048A (en) * | 2018-12-25 | 2019-03-08 | 苏州闪驰数控系统集成有限公司 | A kind of artificial intelligence early warning system |
| CN110751164A (en) * | 2019-03-01 | 2020-02-04 | 西安电子科技大学 | Old man travel abnormity detection method based on location service |
| CN111191511A (en) * | 2019-12-03 | 2020-05-22 | 北京联合大学 | Method and system for identifying dynamic real-time behaviors of prisons |
| CN111291735A (en) * | 2020-04-30 | 2020-06-16 | 华夏天信（北京）智能低碳技术研究院有限公司 | Underground personnel running abnormal behavior detection method based on trajectory analysis |
| CN113011279A (en) * | 2021-02-26 | 2021-06-22 | 清华大学 | Method and device for recognizing mucosa contact action, computer equipment and storage medium |
| CN113080859A (en) * | 2021-03-31 | 2021-07-09 | 安徽正华生物仪器设备有限公司 | Alzheimer's disease early warning system based on daily behavior analysis |
| CN113743380A (en) * | 2021-11-03 | 2021-12-03 | 江苏博子岛智能产业技术研究院有限公司 | Active tracking method based on video image dynamic monitoring |
| CN115183763A (en) * | 2022-09-13 | 2022-10-14 | 南京北新智能科技有限公司 | Personnel map positioning method based on face recognition and grid method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080169929A1 (en) * | 2007-01-12 | 2008-07-17 | Jacob C Albertson | Warning a user about adverse behaviors of others within an environment based on a 3d captured image stream |
| CN102737474A (en) * | 2011-04-07 | 2012-10-17 | 胡茂林 | Monitoring and alarming for abnormal behavior of indoor personnel based on intelligent video |
| CN103118249A (en) * | 2013-03-13 | 2013-05-22 | 胡茂林 | Housebound elder intelligence video surveillance system |
| CN103425971A (en) * | 2013-08-28 | 2013-12-04 | 重庆大学 | Method for monitoring abnormal behaviors of elderly people living alone in family environment |
-
2016
- 2016-05-30 CN CN201610367782.1A patent/CN105975956A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080169929A1 (en) * | 2007-01-12 | 2008-07-17 | Jacob C Albertson | Warning a user about adverse behaviors of others within an environment based on a 3d captured image stream |
| CN102737474A (en) * | 2011-04-07 | 2012-10-17 | 胡茂林 | Monitoring and alarming for abnormal behavior of indoor personnel based on intelligent video |
| CN103118249A (en) * | 2013-03-13 | 2013-05-22 | 胡茂林 | Housebound elder intelligence video surveillance system |
| CN103425971A (en) * | 2013-08-28 | 2013-12-04 | 重庆大学 | Method for monitoring abnormal behaviors of elderly people living alone in family environment |
Non-Patent Citations (1)
| Title |
|---|
| 田国会 等: "基于混合高斯模型和主成分分析的轨迹分析行为识别方法", 《电子学报》 * |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106571014A (en) * | 2016-10-24 | 2017-04-19 | 上海伟赛智能科技有限公司 | Method for identifying abnormal motion in video and system thereof |
| CN106780566B (en) * | 2016-11-11 | 2019-06-21 | 华南智能机器人创新研究院 | A kind of method and system of target following under the robot based on Delta |
| CN106780566A (en) * | 2016-11-11 | 2017-05-31 | 华南智能机器人创新研究院 | A kind of method and system of the target following under the robot based on Delta |
| CN106568445A (en) * | 2016-11-11 | 2017-04-19 | 吉林大学 | Indoor track prediction method based on bidirectional circulation neural network |
| CN106846729A (en) * | 2017-01-12 | 2017-06-13 | 山东大学 | A kind of fall detection method and system based on convolutional neural networks |
| CN106846729B (en) * | 2017-01-12 | 2020-01-21 | 山东大学 | Tumble detection method and system based on convolutional neural network |
| CN107063258A (en) * | 2017-03-07 | 2017-08-18 | 重庆邮电大学 | A kind of mobile robot indoor navigation method based on semantic information |
| CN106971497A (en) * | 2017-05-03 | 2017-07-21 | 成都宏软科技实业有限公司 | A kind of intelligent safety and defence system based on intelligent domestic |
| CN107153871A (en) * | 2017-05-09 | 2017-09-12 | 浙江农林大学 | Fall detection method based on convolutional neural networks and mobile phone sensor data |
| CN107145878A (en) * | 2017-06-01 | 2017-09-08 | 重庆邮电大学 | Old man's anomaly detection method based on deep learning |
| CN107705334A (en) * | 2017-08-25 | 2018-02-16 | 北京图森未来科技有限公司 | A kind of video camera method for detecting abnormality and device |
| CN107705334B (en) * | 2017-08-25 | 2020-08-25 | 北京图森智途科技有限公司 | Camera abnormity detection method and device |
| CN107766852A (en) * | 2017-12-06 | 2018-03-06 | 电子科技大学 | A kind of man-machine mouse track detection method based on convolutional neural networks |
| CN108537105A (en) * | 2018-02-05 | 2018-09-14 | 中国农业大学 | A kind of hazardous act recognition methods under home environment |
| CN108537105B (en) * | 2018-02-05 | 2020-09-25 | 中国农业大学 | Dangerous behavior identification method in home environment |
| CN108229440A (en) * | 2018-02-06 | 2018-06-29 | 北京奥开信息科技有限公司 | One kind is based on Multi-sensor Fusion indoor human body gesture recognition method |
| CN108446594A (en) * | 2018-02-11 | 2018-08-24 | 四川省北青数据技术有限公司 | Emergency reaction ability assessment method based on action recognition |
| CN108446594B (en) * | 2018-02-11 | 2021-08-06 | 四川省北青数据技术有限公司 | Emergency response capability evaluation method based on action recognition |
| CN109308444A (en) * | 2018-07-16 | 2019-02-05 | 重庆大学 | A kind of abnormal behaviour recognition methods under indoor environment |
| CN109002858B (en) * | 2018-07-23 | 2022-01-28 | 合肥工业大学 | Evidence reasoning-based integrated clustering method for user behavior analysis |
| CN109002858A (en) * | 2018-07-23 | 2018-12-14 | 合肥工业大学 | A kind of clustering ensemble method based on evidential reasoning for user behavior analysis |
| CN109447048A (en) * | 2018-12-25 | 2019-03-08 | 苏州闪驰数控系统集成有限公司 | A kind of artificial intelligence early warning system |
| CN109447048B (en) * | 2018-12-25 | 2020-12-25 | 苏州闪驰数控系统集成有限公司 | Artificial intelligence early warning system |
| CN110751164A (en) * | 2019-03-01 | 2020-02-04 | 西安电子科技大学 | Old man travel abnormity detection method based on location service |
| CN110751164B (en) * | 2019-03-01 | 2022-04-12 | 西安电子科技大学 | Old man travel abnormity detection method based on location service |
| CN111191511A (en) * | 2019-12-03 | 2020-05-22 | 北京联合大学 | Method and system for identifying dynamic real-time behaviors of prisons |
| CN111191511B (en) * | 2019-12-03 | 2023-08-18 | 北京联合大学 | Dynamic real-time behavior recognition method and system for prison |
| CN111291735A (en) * | 2020-04-30 | 2020-06-16 | 华夏天信（北京）智能低碳技术研究院有限公司 | Underground personnel running abnormal behavior detection method based on trajectory analysis |
| CN113011279A (en) * | 2021-02-26 | 2021-06-22 | 清华大学 | Method and device for recognizing mucosa contact action, computer equipment and storage medium |
| CN113080859A (en) * | 2021-03-31 | 2021-07-09 | 安徽正华生物仪器设备有限公司 | Alzheimer's disease early warning system based on daily behavior analysis |
| CN113743380A (en) * | 2021-11-03 | 2021-12-03 | 江苏博子岛智能产业技术研究院有限公司 | Active tracking method based on video image dynamic monitoring |
| CN113743380B (en) * | 2021-11-03 | 2022-02-15 | 江苏博子岛智能产业技术研究院有限公司 | Active tracking method based on video image dynamic monitoring |
| CN115183763A (en) * | 2022-09-13 | 2022-10-14 | 南京北新智能科技有限公司 | Personnel map positioning method based on face recognition and grid method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105975956A (en) | Infrared-panorama-pick-up-head-based abnormal behavior identification method of elderly people living alone | |
| Gong et al. | A real-time fire detection method from video with multifeature fusion | |
| CN108216252B (en) | Subway driver vehicle-mounted driving behavior analysis method, vehicle-mounted terminal and system | |
| CN111291699B (en) | Substation personnel behavior identification method based on monitoring video time sequence action positioning and abnormity detection | |
| CN103237201B (en) | A kind of case video analysis method based on socialization mark | |
| Ren et al. | Deep video anomaly detection: Opportunities and challenges | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| CN104318217A (en) | Face recognition information analysis method and system based on distributed cloud computing | |
| CN106022244A (en) | Unsupervised crowd abnormity monitoring and positioning method based on recurrent neural network modeling | |
| CN104717468B (en) | Cluster scene intelligent monitoring method and system based on the classification of cluster track | |
| CN103425971A (en) | Method for monitoring abnormal behaviors of elderly people living alone in family environment | |
| KR20060031832A (en) | A smart visual security system based on real-time behavior analysis and situation cognizance | |
| CN111540152A (en) | Intelligent security early warning system | |
| CN108540752A (en) | The methods, devices and systems that target object in video monitoring is identified | |
| CN103150736A (en) | Camera motion detecting method based on video monitoring | |
| Yang et al. | Detection of non‐suicidal self‐injury based on spatiotemporal features of indoor activities | |
| CN112232211A (en) | Intelligent video monitoring system based on deep learning | |
| CN108256447A (en) | A kind of unmanned plane video analysis method based on deep neural network | |
| CN105227918B (en) | A kind of intelligent control method | |
| CN109766743A (en) | A kind of intelligent bionic policing system | |
| CN104159088A (en) | System and method of remote monitoring of intelligent vehicle | |
| CN108629327A (en) | A kind of demographic method and device based on image procossing | |
| CN107437314A (en) | Personnel's intrusion detection detecting system | |
| CN110930632A (en) | Early warning system based on artificial intelligence | |
| Li et al. | Application research of artificial intelligent technology in substation inspection tour |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160928 |