CN108280435A - A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation - Google Patents
A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation Download PDFInfo
- Publication number
- CN108280435A CN108280435A CN201810074471.5A CN201810074471A CN108280435A CN 108280435 A CN108280435 A CN 108280435A CN 201810074471 A CN201810074471 A CN 201810074471A CN 108280435 A CN108280435 A CN 108280435A
- Authority
- CN
- China
- Prior art keywords
- monitoring system
- pedestrian
- state
- passenger
- attitude estimation
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation, including:Step S1, monitoring system carry out Image Acquisition;Walkway is divided into multiple regions by step S2, monitoring system;Step S3, monitoring system carry out Attitude estimation to pedestrian according to the image of acquisition, the posture of pedestrian are obtained by extracting the framework information of pedestrian;Step S4, monitoring system in corresponding region, carry out multiple image the tracking of multiple target, and carry out behavior judgement according to the movement locus of pedestrian according to the framework information of pedestrian;Step S5, monitoring system are preset with normal pass state and improper prevailing state according to the different postures and different motion track of pedestrian, if monitoring system judges pedestrian and is in improper prevailing state, send out alarm.The present invention can accurately send out alarm in the case where having abnormal behaviour by examination people, while have monitoring swift with judgement, and monitoring range is wide and other effects.
Description
Technical field
The present invention relates to passenger's behavior recognition methods more particularly to a kind of passenger's abnormal behaviours based on human body attitude estimation
Recognition methods.
Background technology
In the prior art, in the numerous occasion, especially certificates checking of the personnel such as customs, airport, railway station, trip is needed
Visitor or customer are lined up one-to-one examination in respective channel, and in examination, other passengers' or customer in the channel drives in the wrong direction, turns over
More, it trails and the abnormal behaviours such as steals into another country and can influence to check order, need that it is identified.Wherein, know for passenger's abnormal behaviour
Other method routine techniques is that the mode of the setting detection zone in single-lens coloured image is identified, and the method is simply led to
It crosses and visual field is divided into abnormal area and normal region on single-lens image, judge that passenger enters exception using background subtraction
Region then illustrates the abnormal behavior of passenger.Currently, background difference method extraction foreground unavoidably exists due to light, camera lens is white
The luggage that the problems such as balance etc. or even passenger carry will also result in interference.
Invention content
The technical problem to be solved in the present invention is, in view of the deficiencies of the prior art, provides one kind and is estimated based on human body attitude
Passenger's abnormal behaviour recognition methods of meter accurately sends out alarm to realize in the case where having abnormal behaviour by examination people
Prompt, while having monitoring swift with judgement, monitoring range is wide and other effects.
In order to solve the above technical problems, the present invention adopts the following technical scheme that.
A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation, this method are based on a monitoring system and realize,
Described method includes following steps:Step S1, the monitoring system carry out Image Acquisition;Step S2, the monitoring system will go
People channel is divided into multiple regions;Step S3, the monitoring system carries out Attitude estimation according to the image of acquisition to pedestrian, and leads to
The framework information of extraction pedestrian is crossed to obtain the posture of pedestrian;Step S4, the monitoring system exist according to the framework information of pedestrian
In corresponding region, the tracking of multiple target is carried out to multiple image, and behavior judgement is carried out according to the movement locus of pedestrian;Step
S5, the monitoring system are preset with normal pass state and improper logical according to the different postures and different motion track of pedestrian
Row state sends out alarm if the monitoring system judges that pedestrian is in improper prevailing state.
Preferably, the monitoring system includes monitoring camera, and the monitoring system is carried out by monitoring camera
Image Acquisition.
Preferably, after the monitoring system acquisition to image, walkway is marked in described image.
Preferably, the improper prevailing state includes that trailing steals into another country state, crosses guardrail state and retrograde state.
Preferably, it is sent out when the monitoring system judges that pedestrian steals into another country state to trail and steals into another country alarm command, work as institute
It states monitoring system and judges that pedestrian is to send out to cross alarm command when crossing guardrail state, when the monitoring system judges pedestrian
To send out retrograde alarm command when retrograde state.
Preferably, after the completion of the alarm command of the monitoring system is sent out, alarm command is removed.
Preferably, in the step S3, the framework information of the monitoring system extraction pedestrian includes shoulder, arm, leg
With head pose information.
It is disclosed by the invention based on human body attitude estimation passenger's abnormal behaviour recognition methods in, the monitoring system according to
The posture of pedestrian and movement locus judge whether pedestrian is in normal pass state, when monitoring system judges the exception of pedestrian
When behavior, alarm can be accurately sent out in time.Compared to existing technologies, the present invention estimates to carry out pedestrian's limb by human body attitude
Body positions, thus will be more accurate to the analysis of foreground target, while present invention monitoring deterministic process is more rapid, monitoring range
It is wider, it is suitably applied in passenger's behavior identification, monitoring system.
Description of the drawings
Fig. 1 is that the present invention is based on passenger's abnormal behaviour recognition methods flow charts that human body attitude is estimated.
Specific implementation mode
The present invention is described in more detail with reference to the accompanying drawings and examples.
The invention discloses a kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation, please refer to Fig. 1, the party
Method is based on a monitoring system and realizes that described method includes following steps:
Step S1, the monitoring system carry out Image Acquisition;
Walkway is divided into multiple regions by step S2, the monitoring system;
Step S3, the monitoring system carry out Attitude estimation according to the image of acquisition to pedestrian, and by extracting pedestrian's
Framework information obtains the posture of pedestrian;
In corresponding region, more mesh are carried out to multiple image according to the framework information of pedestrian for step S4, the monitoring system
Target tracks, and carries out behavior judgement according to the movement locus of pedestrian;
Step S5, the monitoring system are preset with normal pass shape according to the different postures and different motion track of pedestrian
State and improper prevailing state send out alarm if the monitoring system judges that pedestrian is in improper prevailing state.
In the above method, whether the monitoring system judges pedestrian in normal according to the posture of pedestrian and movement locus
Prevailing state can accurately send out alarm in time when monitoring system judges the abnormal behaviour of pedestrian.Compared with prior art and
Speech, the present invention estimate to carry out pedestrian's body fixed position by human body attitude, thus will more accurately, simultaneously to the analysis of foreground target
Present invention monitoring deterministic process is more rapid, monitoring range is wider, is suitably applied in passenger's behavior identification, monitoring system.
In the present embodiment, the monitoring system includes monitoring camera, and the monitoring system is by monitoring camera
Carry out Image Acquisition.
Further, after the monitoring system acquisition to image, walkway is marked in described image.
As a preferred method, the improper prevailing state include trailing steal into another country state, cross guardrail state and
Retrograde state.
In practical application, is sent out when the monitoring system judges that pedestrian steals into another country state to trail and steals into another country alarm command,
It is sent out when the monitoring system judges that pedestrian is to cross guardrail state and crosses alarm command, when the monitoring system is judged
Pedestrian sends out retrograde alarm command when being retrograde state.Further, after the completion of the alarm command of the monitoring system is sent out,
Remove alarm command.
As a preferred method, in the step S3, the framework information of monitoring system extraction pedestrian include shoulder,
Arm, leg and head pose information.
Passenger's abnormal behaviour recognition methods disclosed by the invention based on human body attitude estimation, as shown in Figure 1, its practical is answered
With in the process, following examples are can refer to:
The monitoring system reads in image according to monitoring camera, and transit passage is marked in video image, right
Image carries out pedestrian's Attitude estimation of more people, and framework information is extracted to the posture for obtaining pedestrian.Further according to the framework information of people
The tracking of multiple target is carried out in the multiple image of the region class of delimitation and movement locus carries out behavior judgement.By the pedestrian in channel
It is divided into normal condition and multiple abnormal conditions such as:Trailing steals into another country, crosses guardrail, retrograde etc..When there is abnormal behaviour, alarm
Notice, behavior without exception normal through.Compared to existing technologies, the present invention is more to obtain by carrying out Attitude estimation to more people
Then the 2D framework informations of people carry out the motion track information that multiframe tracking gets more people.Wherein, the pedestrian that the present invention obtains
Trace information is not simple pedestrian track, but the key point information of human body includes shoulder, arm, leg, the dry letter of first-class limb
Breath, when the monitoring system is by the abnormal behaviour of comprehensive descision pedestrian, so that it may can both be directed to human body limb independent analysis,
The characteristics of motion that multiple limbs can be combined again judges whether the behavior is abnormal.Based on the above process, the present invention, which realizes, to be directed to
Passenger's abnormal behaviour is analyzed, and effectively can carry out early warning to criminal activity, is saved supervision human resources, is more effectively easily tieed up
Hold order.
The above is preferred embodiments of the present invention, is not intended to restrict the invention, all technology models in the present invention
Interior done modification, equivalent replacement or improvement etc. are enclosed, should be included in the range of of the invention protect.
Claims (7)
1. a kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation, which is characterized in that this method is based on a monitoring
System realizes that described method includes following steps:
Step S1, the monitoring system carry out Image Acquisition;
Walkway is divided into multiple regions by step S2, the monitoring system;
Step S3, the monitoring system carry out Attitude estimation, and the skeleton by extracting pedestrian according to the image of acquisition to pedestrian
Information obtains the posture of pedestrian;
In corresponding region, multiple target is carried out to multiple image according to the framework information of pedestrian for step S4, the monitoring system
Tracking, and behavior judgement is carried out according to the movement locus of pedestrian;
Step S5, the monitoring system be preset with according to the different postures and different motion track of pedestrian normal pass state and
Improper prevailing state sends out alarm if the monitoring system judges that pedestrian is in improper prevailing state.
2. passenger's abnormal behaviour recognition methods as described in claim 1 based on human body attitude estimation, which is characterized in that described
Monitoring system includes monitoring camera, and the monitoring system carries out Image Acquisition by monitoring camera.
3. passenger's abnormal behaviour recognition methods as described in claim 1 based on human body attitude estimation, which is characterized in that described
After monitoring system acquisition to image, walkway is marked in described image.
4. passenger's abnormal behaviour recognition methods as described in claim 1 based on human body attitude estimation, which is characterized in that described
Improper prevailing state includes that trailing steals into another country state, crosses guardrail state and retrograde state.
5. passenger's abnormal behaviour recognition methods as claimed in claim 4 based on human body attitude estimation, which is characterized in that work as institute
It states monitoring system and judges that pedestrian is to send out to steal into another country alarm command when state is stolen into another country in trailing, when the monitoring system judges pedestrian
Alarm command is crossed to cross to send out when guardrail state, is sent out when the monitoring system judges that pedestrian is retrograde state retrograde
Alarm command.
6. passenger's abnormal behaviour recognition methods as claimed in claim 5 based on human body attitude estimation, which is characterized in that work as institute
State monitoring system alarm command send out after the completion of, remove alarm command.
7. passenger's abnormal behaviour recognition methods as described in claim 1 based on human body attitude estimation, which is characterized in that described
In step S3, the framework information of the monitoring system extraction includes shoulder, arm, leg and the posture information on head.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810074471.5A CN108280435A (en) | 2018-01-25 | 2018-01-25 | A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810074471.5A CN108280435A (en) | 2018-01-25 | 2018-01-25 | A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108280435A true CN108280435A (en) | 2018-07-13 |
Family
ID=62805301
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810074471.5A Pending CN108280435A (en) | 2018-01-25 | 2018-01-25 | A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108280435A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109522793A (en) * | 2018-10-10 | 2019-03-26 | 华南理工大学 | More people's unusual checkings and recognition methods based on machine vision |
CN109784199A (en) * | 2018-12-21 | 2019-05-21 | 深圳云天励飞技术有限公司 | Analysis method of going together and Related product |
CN109815921A (en) * | 2019-01-29 | 2019-05-28 | 北京融链科技有限公司 | The prediction technique and device of the class of activity in hydrogenation stations |
CN109977796A (en) * | 2019-03-06 | 2019-07-05 | 新华三技术有限公司 | Trail current detection method and device |
CN110110710A (en) * | 2019-06-03 | 2019-08-09 | 北京启瞳智能科技有限公司 | A kind of scene abnormality recognition methods, system and intelligent terminal |
CN110135345A (en) * | 2019-05-15 | 2019-08-16 | 武汉纵横智慧城市股份有限公司 | Activity recognition method, apparatus, equipment and storage medium based on deep learning |
CN110378281A (en) * | 2019-07-17 | 2019-10-25 | 青岛科技大学 | Group Activity recognition method based on pseudo- 3D convolutional neural networks |
CN110874910A (en) * | 2018-08-31 | 2020-03-10 | 杭州海康威视数字技术股份有限公司 | Road surface alarm method, device, electronic equipment and readable storage medium |
CN110930568A (en) * | 2019-12-05 | 2020-03-27 | 江苏中云智慧数据科技有限公司 | Video anti-trailing system and method |
CN111144260A (en) * | 2019-12-19 | 2020-05-12 | 北京文安智能技术股份有限公司 | Detection method, device and system of crossing gate |
CN111145533A (en) * | 2019-12-24 | 2020-05-12 | 安徽虹湾信息技术有限公司 | Pedestrian abnormal traffic behavior pattern recognition management and control system based on urban area |
CN111444804A (en) * | 2020-03-19 | 2020-07-24 | 盛视科技股份有限公司 | Human body checking method and system based on gait recognition |
CN113822250A (en) * | 2021-11-23 | 2021-12-21 | 中船(浙江)海洋科技有限公司 | Ship driving abnormal behavior detection method |
CN114010429A (en) * | 2021-11-03 | 2022-02-08 | 河北医科大学第二医院 | Processing device and isolation management system for new crown pneumonia management |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105518744A (en) * | 2015-06-29 | 2016-04-20 | 北京旷视科技有限公司 | Pedestrian re-identification method and equipment |
CN106778655A (en) * | 2016-12-27 | 2017-05-31 | 华侨大学 | A kind of entrance based on human skeleton is trailed and enters detection method |
CN107194360A (en) * | 2017-05-25 | 2017-09-22 | 智慧航安(北京)科技有限公司 | Inversely pass through object identifying method, apparatus and system |
CN107204114A (en) * | 2016-03-18 | 2017-09-26 | 中兴通讯股份有限公司 | A kind of recognition methods of vehicle abnormality behavior and device |
-
2018
- 2018-01-25 CN CN201810074471.5A patent/CN108280435A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105518744A (en) * | 2015-06-29 | 2016-04-20 | 北京旷视科技有限公司 | Pedestrian re-identification method and equipment |
CN107204114A (en) * | 2016-03-18 | 2017-09-26 | 中兴通讯股份有限公司 | A kind of recognition methods of vehicle abnormality behavior and device |
CN106778655A (en) * | 2016-12-27 | 2017-05-31 | 华侨大学 | A kind of entrance based on human skeleton is trailed and enters detection method |
CN107194360A (en) * | 2017-05-25 | 2017-09-22 | 智慧航安(北京)科技有限公司 | Inversely pass through object identifying method, apparatus and system |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110874910A (en) * | 2018-08-31 | 2020-03-10 | 杭州海康威视数字技术股份有限公司 | Road surface alarm method, device, electronic equipment and readable storage medium |
CN110874910B (en) * | 2018-08-31 | 2021-03-23 | 杭州海康威视数字技术股份有限公司 | Road surface alarm method, device, electronic equipment and readable storage medium |
CN109522793A (en) * | 2018-10-10 | 2019-03-26 | 华南理工大学 | More people's unusual checkings and recognition methods based on machine vision |
CN109522793B (en) * | 2018-10-10 | 2021-07-23 | 华南理工大学 | Method for detecting and identifying abnormal behaviors of multiple persons based on machine vision |
CN109784199B (en) * | 2018-12-21 | 2020-11-24 | 深圳云天励飞技术有限公司 | Peer-to-peer analysis method and related product |
CN109784199A (en) * | 2018-12-21 | 2019-05-21 | 深圳云天励飞技术有限公司 | Analysis method of going together and Related product |
CN109815921A (en) * | 2019-01-29 | 2019-05-28 | 北京融链科技有限公司 | The prediction technique and device of the class of activity in hydrogenation stations |
CN109977796A (en) * | 2019-03-06 | 2019-07-05 | 新华三技术有限公司 | Trail current detection method and device |
CN110135345A (en) * | 2019-05-15 | 2019-08-16 | 武汉纵横智慧城市股份有限公司 | Activity recognition method, apparatus, equipment and storage medium based on deep learning |
CN110110710A (en) * | 2019-06-03 | 2019-08-09 | 北京启瞳智能科技有限公司 | A kind of scene abnormality recognition methods, system and intelligent terminal |
CN110378281A (en) * | 2019-07-17 | 2019-10-25 | 青岛科技大学 | Group Activity recognition method based on pseudo- 3D convolutional neural networks |
CN110930568A (en) * | 2019-12-05 | 2020-03-27 | 江苏中云智慧数据科技有限公司 | Video anti-trailing system and method |
CN111144260A (en) * | 2019-12-19 | 2020-05-12 | 北京文安智能技术股份有限公司 | Detection method, device and system of crossing gate |
CN111145533A (en) * | 2019-12-24 | 2020-05-12 | 安徽虹湾信息技术有限公司 | Pedestrian abnormal traffic behavior pattern recognition management and control system based on urban area |
CN111444804A (en) * | 2020-03-19 | 2020-07-24 | 盛视科技股份有限公司 | Human body checking method and system based on gait recognition |
CN114010429A (en) * | 2021-11-03 | 2022-02-08 | 河北医科大学第二医院 | Processing device and isolation management system for new crown pneumonia management |
CN114010429B (en) * | 2021-11-03 | 2023-12-29 | 河北医科大学第二医院 | Processing device for new coronaries pneumonia management and isolation management system |
CN113822250A (en) * | 2021-11-23 | 2021-12-21 | 中船(浙江)海洋科技有限公司 | Ship driving abnormal behavior detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108280435A (en) | A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation | |
CN109522793B (en) | Method for detecting and identifying abnormal behaviors of multiple persons based on machine vision | |
CN106980829B (en) | Abnormal behaviour automatic testing method of fighting based on video analysis | |
CN104691559B (en) | Monitoring station platform safety door and car door interval safe accessory system and its implementation | |
CN106600631A (en) | Multiple target tracking-based passenger flow statistics method | |
CN110188724A (en) | The method and system of safety cap positioning and color identification based on deep learning | |
CN106241533A (en) | Elevator occupant's comprehensive safety intelligent control method based on machine vision | |
CN106241534B (en) | More people's boarding abnormal movement intelligent control methods | |
CN106006266A (en) | Machine vision establishment method applied to elevator safety monitoring | |
AU2015224526B2 (en) | An image management system | |
CN104318578A (en) | Video image analyzing method and system | |
CN103246869A (en) | Crime monitoring method based on face recognition technology and behavior and sound recognition | |
CN108229407A (en) | A kind of behavioral value method and system in video analysis | |
CN106529496B (en) | A kind of method of engine drivers in locomotive depot real-time video fatigue detecting | |
CN106503632A (en) | A kind of escalator intelligent and safe monitoring method based on video analysis | |
CN109145736B (en) | A kind of detection method that the subway station pedestrian based on video analysis inversely walks | |
CN103440475A (en) | Automatic teller machine user face visibility judging system and method | |
CN112016528B (en) | Behavior recognition method and device, electronic equipment and readable storage medium | |
CN107103300B (en) | Off-duty detection method and system | |
CN108932464A (en) | Passenger flow volume statistical method and device | |
CN109508659A (en) | A kind of face identification system and method for crossing | |
CN112071084A (en) | Method and system for judging illegal parking by utilizing deep learning | |
CN113298059A (en) | Pantograph foreign matter intrusion detection method, device, computer equipment, system and storage medium | |
CN110084987A (en) | A kind of foreign matter inspecting system and method towards rail traffic | |
CN109543577A (en) | A kind of fatigue driving detection method for early warning based on facial expression feature |
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 |