CN106372575B - A kind of remote monitoring intelligent Identification of Images management system - Google Patents
A kind of remote monitoring intelligent Identification of Images management system Download PDFInfo
- Publication number
- CN106372575B CN106372575B CN201610702157.8A CN201610702157A CN106372575B CN 106372575 B CN106372575 B CN 106372575B CN 201610702157 A CN201610702157 A CN 201610702157A CN 106372575 B CN106372575 B CN 106372575B
- Authority
- CN
- China
- Prior art keywords
- module
- face
- hard spot
- barrier
- image
- 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.)
- Expired - Fee Related
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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Abstract
The invention discloses a kind of remote monitoring intelligent Identification of Images management systems, including human-machine operation module, image capture module, data processing module, barrier positioning coordinate obtaining module, face 3-D image reconstructed module, face three-dimensionalreconstruction picture depth extraction module, face recognition module and central processing unit.The present invention by kinect depth transducer carry out facial image to be identified depth image data acquisition and three-dimensionalreconstruction with barrier facial image, the identification of face is then completed by the similarity comparison of depth image；Through experiment it is found that completing the three-dimensional reconstruction of target object using the different depth images of Kinect sensor acquisition, it is only necessary to 48s, and very delicate reconstruction effect is obtained, substantially increase the efficiency of recognition of face.
Description
Technical field
The present invention relates to airport face identification systems, and in particular to a kind of remote monitoring intelligent Identification of Images management system.
Background technique
The influence that face is blocked to face identification system is huge, existing barrier, such as barrier, mask, scarf etc.
Be it is common block, the main influence of recognition of face is:
1) it is difficult to collect the largely face picture with barrier and is used for systematic training；
2) barrier varied styles, form color change is big, is difficult to be portrayed with a simple model；
3) barrier can change the statistical nature of image, characteristics of image, to bring identification to interfere to face identification system.
Existing mainstream technology is directed to the method that the face identification system blocked with barrier all uses removal barrier,
The disadvantage is that barrier style is more, variation is big, current removal barrier algorithm perfect can not remove barrier, often give image
New noise is left or brings, to influence subsequent recognition performance.
Summary of the invention
The object of the present invention is to provide a kind of remote monitoring intelligent Identification of Images management systems, realize band barrier
Face quick identification, accuracy is high.
To achieve the above object, the technical scheme adopted by the invention is as follows:
A kind of remote monitoring intelligent Identification of Images management system, including
Human-machine operation module, for inputting various data and control command；
Image capture module, for carrying out the acquisition of video data by camera, and by the collected video data of institute
Data processing module and central processing unit are sent to by data transmission module；
Data processing module, for obtaining depth image by kinect depth transducer, and to the depth image of acquisition
Data are filtered using gaussian pyramid and down-sampling, carry out feature to face and barrier using Haar feature and integrogram
Description by Adaboost algorithm training strong classifier, and uses screening type cascade system, carries out human face region and barrier area
The identification in domain directly initiates alarm module and alarms if barrier region covers entire human face region；
Barrier positions coordinate obtaining module, in place for carrying out barrier institute according to the recognition result of data processing module
The coordinate setting set, and face 3-D image reconstructed module is sent by resulting coordinate setting data；
Face 3-D image reconstructed module, for obtaining resulting human face region and obstacle by kinect depth transducer
The depth image of object area, and the face depth image in the depth image and database of resulting barrier region is carried out
Trigonometric ratio, the depth image that all trigonometric ratios are then merged in scale space, which constructs, is layered Signed Distance Field, in field of adjusting the distance
All voxel applications entirety triangulations generate one and cover the convex closure of all voxels, and utilize
The barrier contour surface of acquisition and face contour surface are pressed obstacle level by MarchingTetrahedra algorithm construction contour surface
The coordinate setting set is spliced, to complete the reconstruct of face 3-D image；
Face three-dimensionalreconstruction picture depth extraction module, for extracting reconstruct three-dimensional face by kinect depth transducer
Depth image, and the depth image that will acquire is sent to face recognition module and is stored；
Face recognition module, for resulting human face region depth image and face three-dimensionalreconstruction picture depth to be extracted mould
The depth image that block obtains compares, and carries out the identification of face, if gap is less than some thresholding, then it is assumed that be otherwise to recognize
Not to be；
Central processing unit carries out alarm module for receiving the recognition result of face recognition module, and according to recognition result
Opening and closing；These orders are sent for receiving the various control commands of man-machine operation module input, and according to preset algorithm
To corresponding module；It is modified for user's registration, rights management and password；It is also used to the data inputted according to human-machine operation module
Call instruction is called corresponding data to be sent to display screen in the database and is shown.
Preferably, the alarm module includes,
Voice alarm module, the control command for being sent according to central processing unit issue audio alert；
Short message editing sending module, control command for being sent according to central processing unit send early warning short message to specified
Mobile terminal.
Preferably, the positioning coordinate uses a revisable hard spot table.
Preferably, the barrier positioning coordinate obtaining module carries out the acquisition of positioning coordinate by following steps:
S1, it is obtained according to human face region and barrier region using face kinetic model of the ADAMS foundation with barrier
ADAMS hard spot file is obtained, the location information of each hard spot of the face with barrier is included at least in ADAMS hard spot file；
S2, the coordinate values for reading each hard spot in ADAMS hard spot file form a revisable hard spot table, hard spot table
In include that each hard spot fix name and the distance between the corresponding coordinate values of each hard spot and two neighboring coordinate are worth.
Preferably, the hard spot table is established by following steps:
It is imported in an EXCEL file using the coordinate values that Matlab reads each hard spot in the ADAMS hard spot file,
There is between each hard spot fix name, coordinate values and two neighboring coordinate storage in first list of EXCEL file
Distance；Hard spot fix name is placed in the first row of the second list of EXCEL file, secondary series is linked in the first list accordingly
Coordinate values, third arranges the distance between corresponding two coordinates for being linked in the first list, and EXCEL file is described
Revisable hard spot table.
Preferably, the camera is infrared camera.
The invention has the following advantages:
The acquisition of the depth image data of facial image to be identified is carried out by kinect depth transducer and with obstacle
Then the three-dimensionalreconstruction of object facial image completes the identification of face by the similarity comparison of depth image；Through experiment it is found that benefit
The three-dimensional reconstruction that target object is completed with the different depth images that Kinect sensor acquires, only needs 48s, and obtain non-
Often fine reconstruction effect, substantially increases the efficiency of recognition of face.
Detailed description of the invention
Fig. 1 is a kind of system block diagram of remote monitoring intelligent Identification of Images management system of the embodiment of the present invention.
Specific embodiment
In order to which objects and advantages of the present invention are more clearly understood, the present invention is carried out with reference to embodiments further
It is described in detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to limit this hair
It is bright.
As shown in Figure 1, the embodiment of the invention provides a kind of remote monitoring intelligent Identification of Images management systems, including
Human-machine operation module, for inputting various data and control command；
Image capture module, for carrying out the acquisition of video data by camera, and by the collected video data of institute
Data processing module and central processing unit are sent to by data transmission module；
Data processing module, for obtaining depth image by kinect depth transducer, and to the depth image of acquisition
Data are filtered using gaussian pyramid and down-sampling, carry out feature to face and barrier using Haar feature and integrogram
Description by Adaboost algorithm training strong classifier, and uses screening type cascade system, carries out human face region and barrier area
The identification in domain directly initiates alarm module and alarms if barrier region covers entire human face region；
Barrier positions coordinate obtaining module, in place for carrying out barrier institute according to the recognition result of data processing module
The coordinate setting set, and face 3-D image reconstructed module is sent by resulting coordinate setting data；
Face 3-D image reconstructed module, for obtaining resulting human face region and obstacle by kinect depth transducer
The depth image of object area, and the face depth image in the depth image and database of resulting barrier region is carried out
Trigonometric ratio, the depth image that all trigonometric ratios are then merged in scale space, which constructs, is layered Signed Distance Field, in field of adjusting the distance
All voxel applications entirety triangulations generate one and cover the convex closure of all voxels, and utilize
The barrier contour surface of acquisition and face contour surface are pressed obstacle level by MarchingTetrahedra algorithm construction contour surface
The coordinate setting set is spliced, to complete the reconstruct of face 3-D image；
Face three-dimensionalreconstruction picture depth extraction module, for extracting reconstruct three-dimensional face by kinect depth transducer
Depth image, and the depth image that will acquire is sent to face recognition module and is stored；
Face recognition module, for resulting human face region depth image and face three-dimensionalreconstruction picture depth to be extracted mould
The depth image that block obtains compares, and carries out the identification of face, if gap is less than some thresholding, then it is assumed that be otherwise to recognize
Not to be；
Central processing unit carries out alarm module for receiving the recognition result of face recognition module, and according to recognition result
Opening and closing；These orders are sent for receiving the various control commands of man-machine operation module input, and according to preset algorithm
To corresponding module；It is modified for user's registration, rights management and password；It is also used to the data inputted according to human-machine operation module
Call instruction is called corresponding data to be sent to display screen in the database and is shown.
The alarm module includes,
Voice alarm module, the control command for being sent according to central processing unit issue audio alert；
Short message editing sending module, control command for being sent according to central processing unit send early warning short message to specified
Mobile terminal.
The positioning coordinate uses a revisable hard spot table.
The barrier positioning coordinate obtaining module carries out the acquisition of positioning coordinate by following steps:
S1, it is obtained according to human face region and barrier region using face kinetic model of the ADAMS foundation with barrier
ADAMS hard spot file is obtained, the location information of each hard spot of the face with barrier is included at least in ADAMS hard spot file；
S2, the coordinate values for reading each hard spot in ADAMS hard spot file form a revisable hard spot table, hard spot table
In include that each hard spot fix name and the distance between the corresponding coordinate values of each hard spot and two neighboring coordinate are worth.
The hard spot table is established by following steps:
It is imported in an EXCEL file using the coordinate values that Matlab reads each hard spot in the ADAMS hard spot file,
There is between each hard spot fix name, coordinate values and two neighboring coordinate storage in first list of EXCEL file
Distance；Hard spot fix name is placed in the first row of the second list of EXCEL file, secondary series is linked in the first list accordingly
Coordinate values, third arranges the distance between corresponding two coordinates for being linked in the first list, and EXCEL file is described
Revisable hard spot table.
The camera is infrared camera.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the principle of the present invention, it can also make several improvements and retouch, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (6)
1. a kind of remote monitoring intelligent Identification of Images management system, which is characterized in that including
Human-machine operation module, for inputting various data and control command；
Image capture module passes through for carrying out the acquisition of video data by camera, and by the collected video data of institute
Data transmission module is sent to data processing module and central processing unit；
Data processing module, for obtaining depth image by kinect depth transducer, and to the depth image data of acquisition
It is filtered using gaussian pyramid and down-sampling, feature is carried out to face and barrier using Haar feature and integrogram and is retouched
It states, by Adaboost algorithm training strong classifier, and uses screening type cascade system, carry out human face region and barrier region
Identification directly initiate alarm module if barrier region covers entire human face region and alarm；
Barrier positions coordinate obtaining module, for carrying out barrier position according to the recognition result of data processing module
Coordinate setting, and face 3-D image reconstructed module is sent by resulting coordinate setting data；
Face 3-D image reconstructed module, for obtaining resulting human face region and barrier area by kinect depth transducer
The depth image in domain, and the face depth image in the depth image and database of resulting barrier region is subjected to triangle
Change, the depth image building layering Signed Distance Field of all trigonometric ratios is then merged in scale space, is owned in field of adjusting the distance
Voxel applications entirety triangulation generate the convex closure for covering all voxels, and utilize Marching
The barrier contour surface of acquisition and face contour surface are pressed the coordinate of Obstacle Position by Tetrahedra algorithm construction contour surface
Positioning is spliced, to complete the reconstruct of face 3-D image；
Face three-dimensionalreconstruction picture depth extraction module, for extracting the depth of reconstruct three-dimensional face by kinect depth transducer
Image is spent, and the depth image that will acquire is sent to face recognition module and is stored；
Face recognition module, for obtaining resulting human face region depth image and face three-dimensionalreconstruction picture depth extraction module
The depth image taken compares, and carries out the identification of face, if gap is less than some thresholding, then it is assumed that be, otherwise it is assumed that not
It is；
Central processing unit carries out opening for alarm module for receiving the recognition result of face recognition module, and according to recognition result
It closes；It sends these orders to pair for receiving the various control commands of man-machine operation module input, and according to preset algorithm
The module answered；It is modified for user's registration, rights management and password；It is also used to the data call inputted according to human-machine operation module
Order is called corresponding data to be sent to display screen in the database and is shown.
2. a kind of remote monitoring intelligent Identification of Images management system according to claim 1, which is characterized in that the alarm
Module includes:
Voice alarm module, the control command for being sent according to central processing unit issue audio alert；
Short message editing sending module, the control command for being sent according to central processing unit send early warning short message to specified movement
Terminal.
3. a kind of remote monitoring intelligent Identification of Images management system according to claim 1, which is characterized in that the positioning
Coordinate uses a revisable hard spot table.
4. a kind of remote monitoring intelligent Identification of Images management system according to claim 1, which is characterized in that the obstacle
Object positions the acquisition that coordinate obtaining module carries out positioning coordinate by following steps:
S1, it is obtained according to human face region and barrier region using face kinetic model of the ADAMS foundation with barrier
ADAMS hard spot file includes at least the location information of each hard spot of the face with barrier in ADAMS hard spot file；
S2, the coordinate values for reading each hard spot in ADAMS hard spot file form a revisable hard spot table, wrap in hard spot table
Include each hard spot fix name and the distance between the corresponding coordinate values of each hard spot and two neighboring coordinate value.
5. a kind of remote monitoring intelligent Identification of Images management system according to claim 4, which is characterized in that the hard spot
Table is established by following steps:
It is imported in an EXCEL file using the coordinate values that Matlab reads each hard spot in the ADAMS hard spot file,
There is between each hard spot fix name, coordinate values and two neighboring coordinate storage in first list of EXCEL file
Distance；Hard spot fix name is placed in the first row of the second list of EXCEL file, secondary series is linked in the first list accordingly
Coordinate values, third arranges the distance between corresponding two coordinates for being linked in the first list, and EXCEL file is described
Revisable hard spot table.
6. a kind of remote monitoring intelligent Identification of Images management system according to claim 1, which is characterized in that the camera shooting
Head is infrared camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610702157.8A CN106372575B (en) | 2016-08-22 | 2016-08-22 | A kind of remote monitoring intelligent Identification of Images management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610702157.8A CN106372575B (en) | 2016-08-22 | 2016-08-22 | A kind of remote monitoring intelligent Identification of Images management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106372575A CN106372575A (en) | 2017-02-01 |
| CN106372575B true CN106372575B (en) | 2019-05-31 |
Family
ID=57878389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610702157.8A Expired - Fee Related CN106372575B (en) | 2016-08-22 | 2016-08-22 | A kind of remote monitoring intelligent Identification of Images management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106372575B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106933786A (en) * | 2017-03-07 | 2017-07-07 | 福建农林大学 | A kind of three-dimensional laser point cloud data rapid voxel processing method |
| CN107247916A (en) * | 2017-04-19 | 2017-10-13 | 广东工业大学 | A kind of three-dimensional face identification method based on Kinect |
| CN107543531B (en) * | 2017-08-13 | 2019-10-11 | 天津职业技术师范大学 | A kind of Robot visual location system |
| CN107491077A (en) * | 2017-09-22 | 2017-12-19 | 衢州职业技术学院 | A kind of indoor floor-sweeping machine device people based on cell phone application control |
| CN108154503A (en) * | 2017-12-13 | 2018-06-12 | 西安交通大学医学院第附属医院 | A kind of leucoderma state of an illness diagnostic system based on image procossing |
| CN108062546B (en) * | 2018-02-11 | 2020-04-07 | 厦门华厦学院 | Computer face emotion recognition system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398886A (en) * | 2008-03-17 | 2009-04-01 | 杭州大清智能技术开发有限公司 | Rapid three-dimensional face identification method based on bi-eye passiveness stereo vision |
| CN104680135A (en) * | 2015-02-09 | 2015-06-03 | 浙江大学 | Three-dimensional human face mark point detection method capable of resisting expression, posture and shielding changes |
| CN105354987A (en) * | 2015-11-26 | 2016-02-24 | 南京工程学院 | Vehicle fatigue driving detection and identity authentication apparatus, and detection method thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8861800B2 (en) * | 2010-07-19 | 2014-10-14 | Carnegie Mellon University | Rapid 3D face reconstruction from a 2D image and methods using such rapid 3D face reconstruction |
| CN105005755B (en) * | 2014-04-25 | 2019-03-29 | 北京邮电大学 | Three-dimensional face identification method and system |
-
2016
- 2016-08-22 CN CN201610702157.8A patent/CN106372575B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398886A (en) * | 2008-03-17 | 2009-04-01 | 杭州大清智能技术开发有限公司 | Rapid three-dimensional face identification method based on bi-eye passiveness stereo vision |
| CN104680135A (en) * | 2015-02-09 | 2015-06-03 | 浙江大学 | Three-dimensional human face mark point detection method capable of resisting expression, posture and shielding changes |
| CN105354987A (en) * | 2015-11-26 | 2016-02-24 | 南京工程学院 | Vehicle fatigue driving detection and identity authentication apparatus, and detection method thereof |
Non-Patent Citations (3)
| Title |
|---|
| 基于Kinect传感器多深度图像融合的物体三维重建;郭连朋 等;《应用光学》;20140930;第35卷(第5期);第811-816页 |
| 基于Kinect的实时人脸识别系统;杨文璐 等;《计算机应用与软件》;20141231(第5期);第64-67、214页 |
| 基于深度自动编码器的机场安检人脸识别系统设计;张宁 等;《计算机测量与控制》;20151231;第23卷(第2期);第644-647页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106372575A (en) | 2017-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106372575B (en) | A kind of remote monitoring intelligent Identification of Images management system | |
| CN106156740B (en) | Civil Aviation Airport terminal face system for rapidly identifying | |
| KR102465532B1 (en) | Method for recognizing an object and apparatus thereof | |
| CN109740516A (en) | A kind of user identification method, device, electronic equipment and storage medium | |
| US9317762B2 (en) | Face recognition using depth based tracking | |
| US10127439B2 (en) | Object recognition method and apparatus | |
| JP2020523665A (en) | Biological detection method and device, electronic device, and storage medium | |
| WO2020151750A1 (en) | Image processing method and device | |
| Raghavendra et al. | Exploring the usefulness of light field cameras for biometrics: An empirical study on face and iris recognition | |
| CN105469356B (en) | Face image processing process and device | |
| CN106033601B (en) | The method and apparatus for detecting abnormal case | |
| CN105426827A (en) | Living body verification method, device and system | |
| CN106295515B (en) | Determine the method and device of the human face region in image | |
| US20170116705A1 (en) | Method for automatic facial impression transformation, recording medium and device for performing the method | |
| KR101308946B1 (en) | Method for reconstructing three dimensional facial shape | |
| KR100896643B1 (en) | Method and system for modeling face in three dimension by means of aam, and apparatus applied to the same | |
| CN110889334A (en) | Personnel intrusion identification method and device | |
| CN109769326B (en) | Light following method, device and equipment | |
| CN105824421A (en) | Multi-modal biological recognition system and method based on holographic projection interactive mode | |
| CN107103299B (en) | People counting method in monitoring video | |
| CN108197585A (en) | Recognition algorithms and device | |
| CN105580050A (en) | Providing control points in images | |
| KR101696007B1 (en) | Method and device for creating 3d montage | |
| Wang et al. | A novel multi-cue integration system for efficient human fall detection | |
| KR101156154B1 (en) | Pose recognition system, medium and server system base on cylinder structure using depth image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190531 Termination date: 20200822 |