CN108288056A - Object features extraction based on large-scale image data, identification device - Google Patents
Object features extraction based on large-scale image data, identification device Download PDFInfo
- Publication number
- CN108288056A CN108288056A CN201810312091.0A CN201810312091A CN108288056A CN 108288056 A CN108288056 A CN 108288056A CN 201810312091 A CN201810312091 A CN 201810312091A CN 108288056 A CN108288056 A CN 108288056A
- Authority
- CN
- China
- Prior art keywords
- image
- camera
- face
- feature
- object features
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 35
- 230000033001 locomotion Effects 0.000 claims description 27
- 230000015572 biosynthetic process Effects 0.000 claims description 9
- 238000003786 synthesis reaction Methods 0.000 claims description 9
- 238000003709 image segmentation Methods 0.000 claims description 7
- 239000004744 fabric Substances 0.000 claims 1
- 238000004458 analytical method Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000000034 method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003012 network analysis Methods 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
A kind of object features extraction based on large-scale image data of the present invention, identification device, including man face image acquiring device, GPU image processors, CPU processor and video terminal receiving device, the man face image acquiring device, GPU image processors, CPU processor is sequentially connected by data line, the CPU processor passes through network connection video terminal receiving device, the camera is provided with multiple, multiple cameras are located at least three orientation, each camera sliding setting, the man face image acquiring device includes sequentially connected camera, video image encoding and decoding module, Face datection capture module and face characteristic extraction module, image enlarging unit is set on the camera, it is provided with convex lens on described image amplifying unit.The present invention is arranged by multiple cameras, can multi-angle acquire human face feature, and by the setting of image enlarging unit, enhanced convenience when people can be made to identify image.
Description
Technical field
The present invention relates to image processing fields more particularly to a kind of object features based on large-scale image data to extract,
Identification device.
Background technology
There are two the development trend of most mainstream, i.e., high Qinghua, intelligence, the faces used at present for field of video monitoring at present
Identification technology has become one of the important support technology of industry-by-industry application and authentication application.Existing recognition of face row
Industry technology uses the comparison analysis system of face and carries out high definition video decoding based on CPU processor for hardware carrier, face is grabbed
It claps, face spy gives the technology that you extract and face characteristic value compares and is applied to product.Common high-definition video monitoring can be shot
Picture carries out recognition of face analysis, and network analysis host obtains head end video stream or video file, solved to the video of acquisition
Code, face recognition algorithms carry out recognition of face to decoded video and capture stingy figure by the face being discharged to, feature you carry
Algorithm is taken to carry out feature extraction to face picture；Face characteristic is compared with database and exports and compares by face alignment algorithm
As a result.
And currently, the extraction for image is because the angle of shooting is excessively single, detected personage is only capable of in a spy
Fixed angle can obtain the preferable image that captures and, it is difficult to by identifying system, be used not once position is slightly staggered
It is convenient.
Invention content
To overcome disadvantages mentioned above, the purpose of the present invention is to provide energy multi-angle acquisition human face feature, improve accurately
Property based on large-scale image data object features extraction, identification device.
The object features extraction that in order to solve the above technical problem, the present invention provides a kind of based on large-scale image data,
Identification device, including man face image acquiring device, GPU image processors, CPU processor and video terminal receiving device, it is described
Man face image acquiring device, GPU image processors, CPU processor are sequentially connected by data line, and the CPU processor passes through net
Network connects video terminal receiving device, and the camera is provided with multiple, and multiple cameras are located at least three orientation, often
A camera sliding setting, the man face image acquiring device includes sequentially connected camera, video image encoding and decoding mould
Image enlarging unit, described image is arranged on the camera in block, Face datection capture module and face characteristic extraction module
Convex lens is provided on amplifying unit.
Advantageous effect the present invention is based on the extraction of the object features of large-scale image data, identification device is facial image
Collector acquires video/image by camera, and image enlarging unit, can by collected video/image, to be amplified
Acquire the spatial dimension of bigger so that enhanced convenience when image recognition；In addition, video image encoding and decoding module adopts camera
The analog signal collected is converted into digital signal；And pass sequentially through Face datection capture module, the progress of face characteristic extraction module
Discriminance analysis is compared finally by CPU processor in the extraction of feature, and the camera in the application has multiple, and is located at
At least three orientation, the feature of the acquisition character facial of the multiple angles of energy are not in because collected personage is not as defined in
Within sweep of the eye, the phenomenon that not acquiring solves the position that detected personage once acquires and is slightly staggered, it is difficult to pass through knowledge
Other system uses inconvenient problem；GPU image processors are introduced, the picture dealt with is more quick.
Preferably, further include memory, the memory is connect with GPU image processors with CPU processor respectively.To
By GPU image processors and CPU processor, treated that video/image stores, and is checked convenient for the later stage.
Preferably, the GPU image processors include image segmentation unit, image comparison unit.Image segmentation unit energy
It is enough camera is collected and multiple junior units are divided by the amplified image of image enlarging unit store, when examining again
After measuring another image, by another image segmentation at several junior units, image comparison unit carries out junior unit lattice twice
It compares, analysis, multiple junior units of storage and collected multiple junior units is compared, there is the wind for reducing False Rate
Danger.
Preferably, the face characteristic extraction module includes sequentially connected image characteristics extraction chip, characteristics of image conjunction
At chip, image generating module；Multiple of described image feature extraction chip for being obtained according to the man face image acquiring device
The movement of the common characteristic point of picture, determines the skimulated motion durection component of each picture, and according to determining skimulated motion side
Directional characteristic classification is carried out to the plurality of pictures；Described image feature synthesis chip is used for according to described image feature extraction
The direction of motion component for the subject that chip determines carries out movement position compensation so that per pictures to each picture
Motion feature point distribution meet described image feature extraction chip progress directional characteristic classification；The image generating module is used
In carrying out the picture and described that directional characteristic classification generates to the plurality of pictures according to described image feature extraction chip
Characteristics of image synthesis chip carries out the picture generated after movement position compensation, synthesizes the motion picture of the subject.With it is existing
There is technology to compare, the present invention provides the knots by image characteristics extraction chip, characteristics of image synthesis chip, image generating module
Structure designs, and subject under motion state cannot accurately be analyzed, to cause by solving image processor in the prior art
The relatively low technical problem of working efficiency, improves the working efficiency of movement detecting images processor, improves movement detecting images
The image processing effect of processor, improves user experience
Preferably, the video terminal receiving device is mobile phone or computer.It can be checked on computer or mobile phone.
The camera for being preferably located at three orientation is moved by two axis robots, increases adopt as far as possible
The possibility of set task face image still has the possibility being detected even if detected personage's head forces down.
Preferably, further include alarm, the alarm is connect with CPU processor.It is identical after comparing analysis, then it sends out
Go out voice alarm.
Description of the drawings
Fig. 1 is embodiment best in the present embodiment.
Specific implementation mode
The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, so that advantages and features of the invention energy
It is easier to be readily appreciated by one skilled in the art, so as to make a clearer definition of the protection scope of the present invention.
Shown in attached drawing 1, the present embodiment is best one embodiment, a kind of object based on large-scale image data
Feature extraction, identification device, including man face image acquiring device 1, GPU image processors 2, CPU processor 3, video terminal receive
Equipment 4 and memory 5, the man face image acquiring device 1, GPU image processors 2, CPU processor 3 by data line successively
Connection, the CPU processor 3 by network connection video terminal receiving device 4, the memory 5 respectively at GPU images
Reason device 2 is connect with CPU processor 3.The video terminal receiving device 4 can be that mobile phone may be computer.The GPU figures
As processor 2 includes image segmentation unit 21, image comparison unit 22.
Wherein, man face image acquiring device 1 is acquiring the face image of personage to be detected, the man face image acquiring device 1
It is carried including sequentially connected camera 11, video image encoding and decoding module 13, Face datection capture module 14 and face characteristic
Modulus block 15 is provided with image enlarging unit 12 on the camera, convex lens is provided on image enlarging unit 12, to will
The image collected is amplified, and is watched convenient for people.
Camera 11 can be collected and is divided by 12 amplified image of image enlarging unit by image segmentation unit 21
Multiple junior units are stored, after detecting another image again, by another image segmentation at several junior units, image
Junior unit lattice twice are compared, analyze by comparison unit 22, by multiple junior units of storage and collected multiple small lists
Member is compared, and has the risk for reducing False Rate.
The object features based on large-scale image data are extracted, identification device can also include alarm 6, the alarm
Device 6 is connect with CPU processor 3.
The camera 11 is provided with multiple, and multiple cameras 11 are located at least three orientation, in the present embodiment
Left side, right side, front side are each provided with a camera 11, high-definition camera 11 may be used, the image collected information is more
It is clear, reduce False Rate.The each sliding of the camera 11 setting, more specific scheme be, can will be positioned at three orientation
The camera 11 moved by two axis robots, and the camera 11 in each orientation is by both sides manipulator where it
Face is moved up and down so that collected head portrait is more comprehensive.
The face characteristic extraction module 15 includes sequentially connected image characteristics extraction chip 15a, characteristics of image synthesis
Chip 15b, image generating module 15c；Described image feature extraction chip 15a according to the man face image acquiring device 1 for obtaining
The movement of the common characteristic point of the plurality of pictures taken, determines the skimulated motion durection component of each picture, and according to determining mould
The quasi- direction of motion carries out directional characteristic classification to the plurality of pictures；Described image feature synthesis chip 15b is used for according to
The direction of motion component for the subject that image characteristics extraction chip 15a is determined carries out movement position benefit to each picture
It repays so that the motion feature point distribution per pictures meets the directional characteristic classification that described image feature extraction chip 15a is carried out；
The image generating module 15c is used to carry out directional characteristic to the plurality of pictures according to described image feature extraction chip 15a
The picture and described image feature synthesis chip 15b that classification generates carry out the picture generated after movement position compensation, synthesize institute
State the motion picture of subject.
The technical concepts and features of embodiment of above only to illustrate the invention, its object is to allow be familiar with technique
People understands present disclosure and is implemented, and it is not intended to limit the scope of the present invention, all according to spirit of that invention
The equivalent change or modification that essence is done should all cover within the scope of the present invention.
Claims (7)
1. a kind of object features based on large-scale image data are extracted, identification device, including man face image acquiring device, GPU figure
As processor, CPU processor and video terminal receiving device, the man face image acquiring device, GPU image processors, at CPU
Reason device is sequentially connected by data line, and the CPU processor passes through network connection video terminal receiving device, it is characterised in that：
The camera is provided with multiple, and multiple cameras are located at least three orientation, each camera sliding setting, institute
State man face image acquiring device include sequentially connected camera, video image encoding and decoding module, Face datection capture module and
Face characteristic extraction module is arranged image enlarging unit on the camera, convex lens is provided on described image amplifying unit.
2. the object features according to claim 1 based on large-scale image data are extracted, identification device, feature exists
In：Further include memory, the memory is connect with GPU image processors with CPU processor respectively.
3. the object features extraction of opportunity large-scale image data according to claim 2, identification device, feature exist
In：The GPU image processors include image segmentation unit, image comparison unit.
4. the object features extraction of opportunity large-scale image data according to claim 2, identification device, feature exist
In：The face characteristic extraction module includes sequentially connected image characteristics extraction chip, characteristics of image synthesis chip, picture life
At module；The common characteristic for the plurality of pictures that described image feature extraction chip is used to be obtained according to the man face image acquiring device
The movement of point, determines the skimulated motion durection component of each picture, and according to determining skimulated motion direction to multiple described figures
Piece carries out directional characteristic classification；Described image feature synthesis chip is used to be determined according to described image feature extraction chip shot
The direction of motion component of object carries out movement position compensation so that the motion feature point minute per pictures to each picture
Cloth meets the directional characteristic classification of described image feature extraction chip progress；The image generating module is used for, according to the figure
As feature extraction chip carries out the picture and described image feature synthesis core that directional characteristic classification generates to the plurality of pictures
Piece carries out the picture generated after movement position compensation, synthesizes the motion picture of the subject.
5. the object features according to claim 1 based on large-scale image data are extracted, identification device, feature exists
In：The video terminal receiving device is mobile phone or computer.
6. the object features according to claim 1 based on large-scale image data are extracted, identification device, feature exists
In：The camera positioned at three orientation is moved by two axis robots.
7. the object features according to claim 1 based on large-scale image data are extracted, identification device, feature exists
In：Further include alarm, the alarm is connect with CPU processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810312091.0A CN108288056A (en) | 2018-04-09 | 2018-04-09 | Object features extraction based on large-scale image data, identification device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810312091.0A CN108288056A (en) | 2018-04-09 | 2018-04-09 | Object features extraction based on large-scale image data, identification device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108288056A true CN108288056A (en) | 2018-07-17 |
Family
ID=62834353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810312091.0A Pending CN108288056A (en) | 2018-04-09 | 2018-04-09 | Object features extraction based on large-scale image data, identification device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108288056A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109558504A (en) * | 2018-11-21 | 2019-04-02 | 扬州市职业大学 | A kind of image processing method |
| CN110674650A (en) * | 2019-08-26 | 2020-01-10 | 深圳市有方科技股份有限公司 | Multi-channel scanning equipment and method for applying same |
| CN112541172A (en) * | 2020-11-27 | 2021-03-23 | 浪潮电子信息产业股份有限公司 | Computer equipment starting method and related device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354780A (en) * | 2007-07-26 | 2009-01-28 | Lg电子株式会社 | Graphic data processing apparatus and method |
| CN204231493U (en) * | 2014-12-03 | 2015-03-25 | 哈尔滨理工大学 | A kind of movement detecting images processor |
| CN104732601A (en) * | 2014-11-19 | 2015-06-24 | 东北大学 | Automatic high-recognition-rate attendance checking device and method based on face recognition technology |
| CN206164722U (en) * | 2016-09-21 | 2017-05-10 | 深圳市泛海三江科技发展有限公司 | Discuss super electronic monitoring system based on face identification |
| CN206388198U (en) * | 2016-12-28 | 2017-08-08 | 深圳市元视科技有限公司 | A kind of recognition of face comparison device based on GPU image processors |
| CN107491775A (en) * | 2017-10-13 | 2017-12-19 | 理光图像技术（上海）有限公司 | Human face in-vivo detection method, device, storage medium and equipment |
| CN208077194U (en) * | 2018-04-09 | 2018-11-09 | 王雅钲 | Object features extraction based on large-scale image data, identification device |
-
2018
- 2018-04-09 CN CN201810312091.0A patent/CN108288056A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354780A (en) * | 2007-07-26 | 2009-01-28 | Lg电子株式会社 | Graphic data processing apparatus and method |
| CN104732601A (en) * | 2014-11-19 | 2015-06-24 | 东北大学 | Automatic high-recognition-rate attendance checking device and method based on face recognition technology |
| CN204231493U (en) * | 2014-12-03 | 2015-03-25 | 哈尔滨理工大学 | A kind of movement detecting images processor |
| CN206164722U (en) * | 2016-09-21 | 2017-05-10 | 深圳市泛海三江科技发展有限公司 | Discuss super electronic monitoring system based on face identification |
| CN206388198U (en) * | 2016-12-28 | 2017-08-08 | 深圳市元视科技有限公司 | A kind of recognition of face comparison device based on GPU image processors |
| CN107491775A (en) * | 2017-10-13 | 2017-12-19 | 理光图像技术（上海）有限公司 | Human face in-vivo detection method, device, storage medium and equipment |
| CN208077194U (en) * | 2018-04-09 | 2018-11-09 | 王雅钲 | Object features extraction based on large-scale image data, identification device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109558504A (en) * | 2018-11-21 | 2019-04-02 | 扬州市职业大学 | A kind of image processing method |
| CN110674650A (en) * | 2019-08-26 | 2020-01-10 | 深圳市有方科技股份有限公司 | Multi-channel scanning equipment and method for applying same |
| CN112541172A (en) * | 2020-11-27 | 2021-03-23 | 浪潮电子信息产业股份有限公司 | Computer equipment starting method and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106228628B (en) | Check-in system, method and device based on face recognition | |
| Shao et al. | Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing | |
| US8660317B2 (en) | Object image detection method and object image detection device for detecting an object image from an input image | |
| CN105809144B (en) | A kind of gesture recognition system and method using movement cutting | |
| CN104933414B (en) | A kind of living body faces detection method based on WLD-TOP | |
| Granger et al. | A comparison of CNN-based face and head detectors for real-time video surveillance applications | |
| TWI439951B (en) | Facial gender identification system and method and computer program products thereof | |
| CN110458063B (en) | Human face living body detection method for preventing video and photo cheating | |
| CN108288056A (en) | Object features extraction based on large-scale image data, identification device | |
| CN108376256A (en) | One kind is based on ARM processing platform dynamic processing face identification systems and its equipment | |
| CN107480586B (en) | Face characteristic point displacement-based biometric photo counterfeit attack detection method | |
| CN109190456B (en) | Multi-feature fusion overlook pedestrian detection method based on aggregated channel features and gray level co-occurrence matrix | |
| US11315360B2 (en) | Live facial recognition system and method | |
| Sharma et al. | Towards a robust, real-time face processing system using CUDA-enabled GPUs | |
| CN113887387A (en) | Ski field target image generation method, system and server | |
| CN112052730B (en) | 3D dynamic portrait identification monitoring equipment and method | |
| Rukhiran et al. | Effecting of environmental conditions to accuracy rates of face recognition based on IoT solution | |
| Singh et al. | Facial recognition using machine learning algorithms on raspberry Pi | |
| CN110222647B (en) | Face in-vivo detection method based on convolutional neural network | |
| Elias et al. | Understanding the gap between 2D and 3D skeleton-based action recognition | |
| CN208077194U (en) | Object features extraction based on large-scale image data, identification device | |
| Tian et al. | Face anti-spoofing by learning polarization cues in a real-world scenario | |
| Song et al. | Face liveliness detection based on texture and color features | |
| CN113591692A (en) | Multi-view identity recognition method | |
| Ma et al. | Detect and track the dynamic deformation human body with the active shape model modified by motion vectors |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200103 Address after: 215000 2nd Floor, Building 14, Xinghua Industrial Park, No. 2 Shuangma Street, Suzhou Industrial Park, Jiangsu Province Applicant after: Suzhou Wide-sighted Intelligent Technology Co., Ltd. Address before: 276100 Room 401, unit 2, 1 Building 9, Tancheng Road, Linyi, Shandong. Applicant before: Wang Yazheng |
|
| TA01 | Transfer of patent application right |