CN106250808A - Double guard detection method based on three-dimensional camera and system - Google Patents

Double guard detection method based on three-dimensional camera and system Download PDF

Info

Publication number
CN106250808A
CN106250808A CN201610387998.4A CN201610387998A CN106250808A CN 106250808 A CN106250808 A CN 106250808A CN 201610387998 A CN201610387998 A CN 201610387998A CN 106250808 A CN106250808 A CN 106250808A
Authority
CN
China
Prior art keywords
image data
double guard
detection method
data stream
duty personnel
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
Application number
CN201610387998.4A
Other languages
Chinese (zh)
Inventor
刘东海
潘今
潘今一
谢晨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SHANGHAI EUTROVISION SYSTEMS Inc
Original Assignee
SHANGHAI EUTROVISION SYSTEMS Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by SHANGHAI EUTROVISION SYSTEMS Inc filed Critical SHANGHAI EUTROVISION SYSTEMS Inc
Priority to CN201610387998.4A priority Critical patent/CN106250808A/en
Publication of CN106250808A publication Critical patent/CN106250808A/en
Pending legal-status Critical Current

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
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Alarm Systems (AREA)

Abstract

The invention discloses a kind of Double guard detection method based on three-dimensional camera and system, detection method comprises the following steps: obtain depth image data stream and the obtaining step of color image data stream by multiple three-dimensional camera；The depth image data stream got and color image data circulation are turned to the step of converting of the three-dimensional coordinate in system spatial coordinates system of the multiple joints in skeleton；Based on the three-dimensional coordinate in the space coordinates that step of converting is converted, it is analyzed by Double guard computing module, the number of detection duty personnel and posture, it is judged that the judgement step that duty personnel are the most abnormal；And be judged as when duty personnel have abnormal in judging step, then the alarming step that alarming result is backed up by alarm signal simultaneously is sent to master control room.According to the present invention, can determine that the number of the duty personnel of each prison prison room and residing posture state thereof, process in time.

Description

Double guard detection method based on three-dimensional camera and system
Technical field
The present invention relates to a kind of detection method, particularly relate to a kind of Double guard detection method based on three-dimensional camera and System.
Background technology
At night and lunch break, the supervision unit people's police on duty such as detention house are fewer, in order to avoid sending out of accident Raw, two convicts can be arranged to carry out duty in each prison room, when the indoor generation abnormal conditions of prison, it can be at the very first time Reason, and report master control room people's police on duty.But, the duty time mostly typically be night and people fewer when, duty personnel hold Easily leave the post and sleep the events such as hilllock, causing it can not effectively fulfil duty responsibility.In order to effectively supervise indoor duty personnel Duty, typically now use monitoring camera duty personnel are monitored, if looked into by master control room people's police on duty completely See monitor video, owing to prison room is too many, it will be caused to pinpoint the problems in time.
It is currently, there are some solutions, but these solutions too depend on the auxiliary materials such as clothing, when auxiliary Help medicated cap or waistcoat to take off to put on the ground, state on duty can be detected as equally, cause wrong report, increase the work of people's police on duty Amount.If the couchant sleep of duty personnel, or by situations such as wall sleeps, the effect of supervision on duty can be affected.
Summary of the invention
With in appending claims, the supplementary features of the present invention will be described below.In this respect, specifically Before at least one embodiment of the bright present invention, it should be appreciated that the present invention does not limit its application to the thin of structure Joint and in the following description elaboration or the layout of parts illustrated in the accompanying drawings.The present invention can be real with other embodiments Existing, and implement in many ways and carry out.However, it should be understood that the purpose of phraseology and terminology employed herein is only It is for describing, should not be regarded as limiting.
The technical problem to be solved is a kind of Double guard detection method, and a set of Double guard is real-time Detecting system.By Intellectual Analysis Technology based on three-dimensional video information in Double guard detection method, determine that each prison is supervised The number of the duty personnel of room and residing posture state thereof.When the indoor duty personnel of prison leave the post and sleep hilllock when, automatically Externally reporting to the police, the people's police on duty of notice master control room so that it is can process in time, solitary abnormal conditions occur.
The present invention solves above-mentioned technical problem by following technical proposals: a kind of double value based on three-dimensional camera Hilllock detection method, it is characterised in that it comprises the following steps:
Depth image data stream and the obtaining step of color image data stream is obtained by multiple three-dimensional camera；
Multiple joints that the depth image data stream got and color image data circulation turn in skeleton are existed The step of converting of the three-dimensional coordinate in system spatial coordinates system；
Based on the three-dimensional coordinate in the space coordinates that step of converting is converted, carried out point by Double guard computing module Analysis, the number of detection duty personnel and posture, it is judged that the judgement step that duty personnel are the most abnormal；And
It is judged as when duty personnel have abnormal in judging step, then sends alarm signal to master control room and warning is tied simultaneously The alarming step of fruit backup.
Preferably, in described judgement step, the Euclidean distance being determined by between each articulare differentiates with angle The posture of duty personnel.Use the method identify posture, discrimination is higher, and can extend at any time maneuver library meet judge not Demand with posture.
Understand the present invention and attendant advantages thereof in order to conceptual, see accompanying drawing and the side of being preferable to carry out of the shown present invention Content described by formula.By the explanation of (multiple) preferred implementation, in conjunction with the accompanying drawing of exemplary forms, illustrate the present invention Principle, thus other features and advantages of the present invention are apparent from.
Accompanying drawing explanation
See exemplary embodiment described further below, it is an advantage of the invention that it will be apparent that this explanation should Consider the most together:
Fig. 1 is skeleton three-dimensional (3 D) manikin structural representation.
Fig. 2 be the human body simplified just facing to camera, or the back side is facing to people's body node schematic diagram of camera.
Fig. 3 is the human body Node distribution schematic diagram that the number of people side simplified faces toward camera.
Fig. 4 is the skeleton node schematic diagram of sitting posture gesture.
Fig. 5 is Double guard detecting system flow chart.
Fig. 6 is human posture's identifying processing flow chart.
Detailed description of the invention
Several embodiments of the invention described in detail below.By the way of explanation, these embodiments are described, Should not be interpreted as limiting the scope of the present invention inadequately.It is true that those of ordinary skill in the art it will be appreciated that Read this specification and viewing accompanying drawing is understood that, The present invention gives many modes of texturing and the teaching of change, and do not taking off In the case of scope and spirit of the present invention, the present invention can use, use and manufacture many modes of texturing.
As it is shown in figure 5, the invention discloses a kind of Double guard detection method based on three-dimensional camera, including following step Suddenly.
In step S101, obtain depth image data stream and color image data stream by multiple three-dimensional camera.It is preferably Each camera is separate, and to reduce the burden of network transmission, the most real-time result is uploaded, but is not limited to this.
In step S102, the depth image data stream got and color image data are circulated turn to people in step S101 The multiple joints in body skeleton three-dimensional coordinate in system spatial coordinates system.The quantity in joint is preferably 20, but does not limit In this.The skeleton three-dimensional (3 D) manikin structure used in the present embodiment is as it is shown in figure 1, each round dot represents a joint.
Fig. 4 shows the skeleton node schematic diagram of sitting posture gesture.Concrete transform mode has kinds of schemes, such as human body Skeleton identification, skeleton simplification etc., figure 6 illustrates exemplary transform mode, i.e. by skeleton identification and skeleton Simplify the multiple joints turning in skeleton that depth image data stream and color image data circulated to sit at system space Three-dimensional coordinate in mark system.
Three-dimensional coordinate in the space coordinates that step S103 is converted based on step S102, calculates mould by Double guard Block is analyzed, the number of detection duty personnel and posture, it is judged that the judgement step that duty personnel are the most abnormal.
Fig. 6 shows exemplary human posture's identifying processing flow chart.
As shown in Figure 6, the video stream data got multiple joints in being converted into skeleton are sat at system space After three-dimensional coordinate in mark system, identified by human posture, human posture judges by the way of mating.
Fig. 2 shows the node schematic diagram during human body straight-on camera of simplification, and Fig. 3 shows that the human body of simplification is back to camera Time node schematic diagram.
Certainly, straight-on camera and the situation back to camera allow to be reduced to a kind of situation, the most all use the joint shown in Fig. 2 Point schematic diagram.Now the node schematic diagram in the corresponding side situation to camera, i.e. Fig. 2 and Fig. 3 of the node schematic diagram in Fig. 3 is total Altogether can corresponding 3 kinds of situations.
Now, in order to ensure the accuracy of gesture recognition, can come by calculating the angle between some articulare line Carry out the identification of given pose.Utilize the Euclidean distance between certain two articulare and the cosine law, two articulares can be drawn Between angle.Assume to have an x and some y, then the Euclidean distance between these 2 is just:
d ( x , y ) = Σ ( x i - y i ) 2 , i = ( 0 , 1 , 2 , 3 ... n )
If sitting posture gesture to be judged, as described in Figure 5.Upper body lies substantially in straight line, it is allowed within certain limit Bending.Lower limb are clearly present certain angular relationship.According to the geometrical relationship of each node, utilize the Europe between certain two articulare Family name's distance and the cosine law.Correlation formula is as follows:
a = ( x 1 - x 2 ) 2 + ( y 1 - y 2 ) 2 + ( z 1 - z 2 ) 2
b = ( x 2 - x 3 ) 2 + ( y 2 - y 3 ) 2 + ( z 2 - z 3 ) 2
c = ( x 1 - x 3 ) 2 + ( y 1 - y 3 ) 2 + ( z 1 - z 3 ) 2
α = a r c c o s b 2 + c 2 - a 2 2 b c
In step S104, it is judged as in step s 103 when duty personnel have abnormal, then sending alarm signal to master control room Alarming result is backed up simultaneously.Warning can use sound, light and any alarm form of electricity isotype, is reporting to the police it to master control room After, people's police on duty just can process in time.Alarm picture and warning video such as can be stored in clothes by the backup of alarming result Business device data base, it is simple to query history record when that later user needing, and generate various statistics according to historical record data Form, it is simple to analyze.
Finally terminate whole flow process.
Understand according to above-mentioned, in the real-time system combining Double guard detection method based on three-dimensional camera and design, By using data processing equipment to process human posture's state of three-dimensional camera capture.And human posture's state is made one Action sequence, and posture sequence is placed in Double guard, it is possible to analyze the situation of duty personnel, general judgment value in real time Whether hilllock personnel leave the post, and sleep the abnormal conditions such as hilllock, and alarm, the people's police on duty of notice master control room so that it is can and crucial point reason, Stop abnormal conditions to occur.Reduce the wrong report that there is no need simultaneously, alleviate the process workload of people's police.
The invention also discloses a kind of Double guard detecting system based on three-dimensional camera, including:
Acquiring unit, for obtaining depth image data stream and color image data stream by multiple three-dimensional camera；
Conversion unit, turns to people for the depth image data stream got by acquiring unit and color image data circulation The multiple joints in body skeleton three-dimensional coordinate in system spatial coordinates system；
Judging unit, the three-dimensional coordinate in the space coordinates converted based on conversion unit, by Double guard meter Calculation module is analyzed, the number of detection duty personnel and posture, it is judged that duty personnel are the most abnormal；And
Alarm unit, for being judged as when duty personnel have abnormal, then sending alarm signal to master control room in judging step Alarming result is backed up number simultaneously.
The each method embodiment of the present invention all can realize in modes such as software, hardware, firmwares.No matter the present invention be with Software, hardware or firmware mode realize, and instruction code may be stored in the addressable memorizer of any kind of computer In (the most permanent or revisable, volatibility or non-volatile, solid-state or non-solid, fixing or Removable medium of person etc.).Equally, memorizer can e.g. programmable logic array (Programmable Array Logic, is called for short " PAL "), random access memory (Random Access Memory, be called for short " RAM "), able to programme read-only deposit Reservoir (Programmable Read Only Memory is called for short " PROM "), read only memory (Read-Only Memory, letter Claim " ROM "), Electrically Erasable Read Only Memory (Electrically Erasable Programmable ROM, be called for short " EEPROM "), disk, CD, digital versatile disc (Digital Versatile Disc, be called for short " DVD ") etc..
It should be noted that in the claim and description of this patent, the relation of such as first and second or the like Term is used merely to separate an entity or operation with another entity or operating space, and not necessarily requires or imply Relation or the order of any this reality is there is between these entities or operation.And, term " includes ", " comprising " or its Any other variant is intended to comprising of nonexcludability so that include the process of a series of key element, method, article or Equipment not only includes those key elements, but also includes other key elements being not expressly set out, or also include for this process, The key element that method, article or equipment are intrinsic.In the case of not having more restriction, statement " include one " and limit wants Element, it is not excluded that there is also other identical element in including the process of described key element, method, article or equipment.
Although by referring to some of the preferred embodiment of the invention, the present invention being shown and described, but It will be understood by those skilled in the art that can to it, various changes can be made in the form and details, without departing from this Bright spirit and scope.

Claims (9)

1. a Double guard detection method based on three-dimensional camera, it is characterised in that comprise the following steps:
Depth image data stream and the obtaining step of color image data stream is obtained by multiple three-dimensional camera；
The depth image data stream got and color image data are circulated the multiple joints turning in skeleton in system The step of converting of the three-dimensional coordinate in space coordinates；
Based on the three-dimensional coordinate in the space coordinates that step of converting is converted, it is analyzed by Double guard computing module, inspection The number of measured value hilllock personnel and posture, it is judged that the judgement step that duty personnel are the most abnormal；And
It is judged as when duty personnel have abnormal in judging step, then sends alarm signal to master control room simultaneously that alarming result is standby The alarming step of part.
Double guard detection method the most according to claim 1, it is characterised in that
In described judgement step, the Euclidean distance being determined by between each articulare and angle are to differentiate the appearance of duty personnel Gesture.
Double guard detection method the most according to claim 1, it is characterised in that
The depth image data stream and the color image data stream front end embedded processing platform that get in obtaining step enter Row transmission.
Double guard detection method the most according to claim 1, it is characterised in that
Each camera is separate.
Double guard detection method the most according to claim 1, it is characterised in that
Include that skeleton identification step and skeleton simplify step at step of converting.
Double guard detection method the most according to claim 1, it is characterised in that
Judging that step includes that human posture's identification step and human posture mate step.
Double guard detection method the most according to claim 1, it is characterised in that
In alarming step with sound, optical, electrical in any one mode signal an alert.
Double guard detection method the most according to claim 1, it is characterised in that
Alarming result is backed up in the way of data base's storage by alarming step.
9. a Double guard detecting system based on three-dimensional camera, it is characterised in that including:
Acquiring unit, for obtaining depth image data stream and color image data stream by multiple three-dimensional camera；
Conversion unit, turns to human bone for the depth image data stream got by acquiring unit and color image data circulation The multiple joints in bone three-dimensional coordinate in system spatial coordinates system；
Judging unit, the three-dimensional coordinate in the space coordinates converted based on conversion unit, calculate mould by Double guard Block is analyzed, the number of detection duty personnel and posture, it is judged that duty personnel are the most abnormal；And
Alarm unit, for being judged as when duty personnel have abnormal in judging step, then sends alarm signal same to master control room Time by alarming result back up.
CN201610387998.4A 2016-06-02 2016-06-02 Double guard detection method based on three-dimensional camera and system Pending CN106250808A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610387998.4A CN106250808A (en) 2016-06-02 2016-06-02 Double guard detection method based on three-dimensional camera and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610387998.4A CN106250808A (en) 2016-06-02 2016-06-02 Double guard detection method based on three-dimensional camera and system

Publications (1)

Publication Number Publication Date
CN106250808A true CN106250808A (en) 2016-12-21

Family

ID=57613196

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610387998.4A Pending CN106250808A (en) 2016-06-02 2016-06-02 Double guard detection method based on three-dimensional camera and system

Country Status (1)

Country Link
CN (1) CN106250808A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
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
CN111314854A (en) * 2020-01-15 2020-06-19 秦皇岛泰和安科技有限公司 Method for monitoring on-duty personnel and monitoring system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102855470A (en) * 2012-07-31 2013-01-02 中国科学院自动化研究所 Estimation method of human posture based on depth image
CN103366565A (en) * 2013-06-21 2013-10-23 浙江理工大学 Method and system of detecting pedestrian running red light based on Kinect
CN103455657A (en) * 2013-06-21 2013-12-18 浙江理工大学 Kinect based field operation simulation method and Kinect based field operation simulation system
CN104021538A (en) * 2013-02-28 2014-09-03 株式会社理光 Object positioning method and device
CN104157107A (en) * 2014-07-24 2014-11-19 燕山大学 Human body posture correction device based on Kinect sensor
CN104182987A (en) * 2013-05-22 2014-12-03 株式会社日立制作所 People counting device and people trajectory analysis device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102855470A (en) * 2012-07-31 2013-01-02 中国科学院自动化研究所 Estimation method of human posture based on depth image
CN104021538A (en) * 2013-02-28 2014-09-03 株式会社理光 Object positioning method and device
CN104182987A (en) * 2013-05-22 2014-12-03 株式会社日立制作所 People counting device and people trajectory analysis device
CN103366565A (en) * 2013-06-21 2013-10-23 浙江理工大学 Method and system of detecting pedestrian running red light based on Kinect
CN103455657A (en) * 2013-06-21 2013-12-18 浙江理工大学 Kinect based field operation simulation method and Kinect based field operation simulation system
CN104157107A (en) * 2014-07-24 2014-11-19 燕山大学 Human body posture correction device based on Kinect sensor

Cited By (3)

* Cited by examiner, † Cited by third party
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
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
CN111314854A (en) * 2020-01-15 2020-06-19 秦皇岛泰和安科技有限公司 Method for monitoring on-duty personnel and monitoring system

Similar Documents

Publication Publication Date Title
CN102859565B (en) Method and system for security system tampering detection
Leone et al. Detecting falls with 3D range camera in ambient assisted living applications: A preliminary study
JP5143212B2 (en) Method and apparatus for modeling behavior using probability distribution functions
CN102811343B (en) Intelligent video monitoring system based on behavior recognition
KR101536410B1 (en) Apparatus and method for managing of data in energy management system
Zouba et al. Monitoring activities of daily living (ADLs) of elderly based on 3D key human postures
CN109872491A (en) Fire monitoring method, device, electronic equipment and system
CN109543607A (en) Object abnormal state detection method, system, monitor system and storage medium
CN103514694A (en) Intrusion detection monitoring system
KR102161244B1 (en) Industrial equipment monitoring and alarm apparatus and its method
JP2018072881A (en) Fire disaster monitoring system
Zouba et al. Multisensor fusion for monitoring elderly activities at home
Mousse et al. Percentage of human-occupied areas for fall detection from two views
CN112929611A (en) Campus intelligent security prevention and control system
CN209088626U (en) Intelligent substation patrol monitoring system
CN106250808A (en) Double guard detection method based on three-dimensional camera and system
CN113313899A (en) Security monitoring analysis processing method based on big data
JP4111660B2 (en) Fire detection equipment
CN102610054A (en) Video-based getting up detection system
CN111191499A (en) Fall detection method and device based on minimum center line
CN207530963U (en) A kind of illegal geofence system based on video monitoring
CN115690914A (en) Abnormal behavior reminding method and device, electronic equipment and storage medium
CN115841730A (en) Video monitoring system and abnormal event detection method
CN111753587A (en) Method and device for detecting falling to ground
CN108551493A (en) Smart city environment monitoring device and method based on mobile cloud computing

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20161221

RJ01 Rejection of invention patent application after publication