CN107194357A - Face identification system and its detection method - Google Patents
Face identification system and its detection method Download PDFInfo
- Publication number
- CN107194357A CN107194357A CN201710380058.7A CN201710380058A CN107194357A CN 107194357 A CN107194357 A CN 107194357A CN 201710380058 A CN201710380058 A CN 201710380058A CN 107194357 A CN107194357 A CN 107194357A
- Authority
- CN
- China
- Prior art keywords
- face
- identification system
- module
- characteristic
- detection
- 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
- G06V40/10—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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a kind of face identification system and its detection method.The face identification system includes panel end, face detection module, demographic data storehouse, face recognition module, the panel end continuous collecting face source data, and sends to the face detection module, and the demographic data storehouse stores personnel's essential information.The detection method of the face identification system includes the steps such as pretreatment, Local Characteristic analysis, Dynamic Local specificity analysis, interative computation.The face recognition technology that the face identification system and its detection method that the present invention is provided solve prior art is present to light environment sensitivity, and big technical problem is influenceed by shelter.
Description
Technical field
The present invention relates to living things feature recognition field, and in particular to a kind of face identification system and its detection method.
Background technology
The technology that face recognition technology is a kind of association by facial characteristics and between them to be identified.Mostly
The image or image of face are absorbed by camera, catches and records core point information therein, the processing and conversion passed through
Data signal is changed into, it is final to produce matching rate signal.
Face recognition technology possesses contactless, concurrency, the low advantage of hardware device requirement.But there is also to week simultaneously
The light environment enclosed is sensitive, the problem of shelter such as hair, jewelry has considerable influence to the degree of accuracy.
The content of the invention
Exist to solve the face recognition technology of prior art to light environment sensitivity, influenceed big technology to ask by shelter
Topic, the present invention provides a kind of face identification system solved the above problems and its detection method.
A kind of face identification system, including panel end, face detection module, demographic data storehouse, face recognition module, it is described
Panel end continuous collecting face source data, and send to the face detection module, the demographic data storehouse stores personnel's base
This information;
When carrying out user's registration operation, the face detection module detects face characteristic in the face source data, will
It is created as master pattern, and sends to demographic data storehouse storage, is matched with personnel's essential information;
When carrying out user's identification operation, the face detection module detects face characteristic in the face source data, will
It is created as detection model, and sends to the face recognition module, and the face recognition module transfers the demographic data storehouse
The interior master pattern is matched with the detection model, obtains matching rate.
In a kind of preferred embodiment for the face identification system that the present invention is provided, the face identification system also includes people
Face template database, the face template database stores the basic model of common face;
When carrying out user's registration operation, the face detection module detects face characteristic in the face source data, and
The suitable basic model is selected in the face template database based on the face characteristic, the face characteristic is filled out
Fill in the basic model, so as to be created that the master pattern, the master pattern is sent to the demographic data storehouse and stored up
Deposit, matched with personnel's essential information;
When carrying out user's identification operation, the face detection module is similarly created that the detection model, by the detection mould
Type is sent to the face recognition module, and the face recognition module transfers the master pattern and detection model progress
Match somebody with somebody, obtain matching rate.
In a kind of preferred embodiment of face identification system that the present invention is provided, the face identification system also include with
Track module;
The matching rate of the acquisition of the face recognition module is temporarily stored into the tracking module, and the tracking module drives described
Face detection module creates another detection model again, and the face recognition module is again to detection model progress
Match somebody with somebody, obtain another matching rate;
Repeat after certain number of times, the tracking module exports recognition result with reference to multiple matching rates to the panel end.
In a kind of preferred embodiment for the face identification system that the present invention is provided, the panel end includes data acquisition list
Member, switch unit, display unit, the data acquisition unit continuous collecting face source data, and send to the Face datection
Module;
During user's registration operation to be carried out, the face detection module is switched to by login state by the switch unit, will
The face characteristic is created as the master pattern;
During user's identification operation to be carried out, the face detection module is switched to by identification state by the switch unit, will
The face characteristic is created as the detection model;
The tracking module exports the recognition result to the display unit.
In a kind of preferred embodiment for the face identification system that the present invention is provided, the detection side of the face identification system
Method, comprises the following steps:
Step one:Pretreatment, detects the edge of portrait in the face source data, removes the high intensity in the face source data
Noise, extract shape of face feature and texture feature, be designated as foundation characteristic;
Step 2:Local Characteristic is analyzed:Set and eliminate edge effect, the induction coefficient of light environment error, be according to the sensing
Based on number, part edge effect and light environment error in the foundation characteristic are shielded, effective detection result is obtained, is designated as face
Portion's feature;
Step 3:Dynamic Local specificity analysis:Induction coefficient described in the step 2 is changed, another face is obtained special
Levy;
Step 4:Interative computation:Preferably one facial characteristics is chosen as the foundation characteristic, progress institute of reruning
Step 2 and the step 3 are stated, until obtaining the suitable facial characteristics is used as the face characteristic.
Compared to prior art, the face identification system that the present invention is provided has the advantages that:
First, the face identification system is provided with the face detection module, and edge effect, light environment are mitigated using interative computation
Influence of the error to recognition accuracy, obtains accurate face characteristic, between recognition speed and recognition accuracy, reaches
Optimal matching effect.
2nd, the face identification system is provided with the face template database, sets up the master pattern or the detection
The basic model therein is based on during model, modeling efficiency is both substantially increased, and effectively speeded identification versus speed.
The speed of service for making the face identification system overall is greatly promoted.
3rd, the face identification system is provided with the tracking module, and rerun the multinomial matching rate of acquisition, effectively subtracts
The influence for the error that light luminous environment and overcover are produced to identification, so as to obtain optimal recognition accuracy.
Brief description of the drawings
Fig. 1 is the structural representation for the face identification system that the present invention is provided.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.
Referring to Fig. 1, being the structural representation for the face identification system 1 that the present invention is provided.
The face identification system 1 includes panel end 11, face detection module 12, face template database 13, personnel's number
According to storehouse 14, face recognition module 15, tracking module 16.The panel end 11 include data acquisition unit 111, switch unit 112,
Display unit 113.
The continuous collecting face image data of data acquisition unit 111, and send to the face detection module 12.Institute
The basic model that face template database 13 stores common face is stated, the demographic data storehouse 14 stores the base of related personnel
This information.
During user's registration operation to be carried out, the switch unit 112 switches the face detection module 12 to login state,
The processing that described 12 pairs of face image datas of face detection module are followed the steps below successively:
Step one:Pretreatment, detects the edge of portrait in the face image data, removes the height in the face image data
The noise of intensity, extracts shape of face feature and texture feature, is designated as foundation characteristic;
Step 2:Local Characteristic is analyzed:Set and eliminate edge effect, the induction coefficient of light environment error, be according to the sensing
Based on number, part edge effect and light environment error in the foundation characteristic are shielded, the effective detection result of acquisition is designated as
Facial characteristics;
Step 3:Dynamic Local specificity analysis:Induction coefficient described in the step 2 is changed, another face is obtained special
Levy;
Step 4:Interative computation:Preferably one facial characteristics is chosen as the foundation characteristic, progress institute of reruning
State step 2 and the step 3 totally ten times, obtain the sufficiently exact facial characteristics and be used as face characteristic.
Step 5:Create master pattern:Based on the face characteristic, a conjunction is chosen in the face template database 13
The suitable basic model, the basic model is filled in by the face characteristic, so as to be created that master pattern.
The master pattern is sent to demographic data storehouse storage 14, with the personnel's essential information wherein stored
Matched, so as to realize the registration of the face information of the user.
During user's identification operation to be carried out, the switch unit 112 switches the face detection module 12 to identification state,
The processing that described 12 pairs of face image datas of face detection module are followed the steps below successively:
Wherein, step one to step 4 with carry out user's registration operation when it is consistent, therefore not to repeat here.
Step 5:Create detection model:Based on the face characteristic, a conjunction is chosen in the face template database 13
The suitable basic model, the basic model is filled in by the face characteristic, so as to be created that detection model.
The detection model is sent to the face recognition module 15.The face recognition module 15 transfers the personnel
With the master pattern of the detection model based on the same basic model in database storage 14, contrasted one by one.
The matching rate for choosing wherein highest one is sent to the tracking module 16.
The tracking module 16 drives 12 pairs of the face detection module data acquisition unit 111 newly to gather the people
Face image data carry out single treatment again, create another detection model, send to the face recognition module 15, and same
The master pattern is contrasted.The tracking module 16 reruns five times, obtains totally six matching rates.
The tracking module 16 determines final matching results according to six matching rates, and sends to the display unit
113 displays.
Compared to prior art, the face identification system 1 that the present invention is provided has the advantages that:
First, the face identification system 1 mitigates edge effect, light environment error to the shadow of recognition accuracy using interative computation
Ring, obtain accurate face characteristic.Such a detection mode has reached most preferably between recognition speed and recognition accuracy
Matching effect.
2nd, the face identification system 1 is provided with the face template database 13, sets up the master pattern or the inspection
The basic model therein is based on when surveying model, modeling efficiency is both substantially increased, and has effectively speeded identification contrast speed
Degree.The speed of service for making the face identification system 1 overall is greatly promoted.
3rd, the face identification system 1 is provided with the tracking module 16, and rerun the multinomial matching rate of acquisition, has
Effect alleviates the influence for the error that luminous environment and overcover are produced to identification, so as to obtain optimal recognition accuracy.
Embodiments of the invention are the foregoing is only, are not intended to limit the scope of the invention, it is every to utilize this hair
Equivalent structure or equivalent flow conversion that bright description is made, or directly or indirectly it is used in other related technology necks
Within domain, the scope of patent protection for being similarly included in the present invention.
Claims (5)
1. a kind of face identification system, it is characterised in that:Including panel end, face detection module, demographic data storehouse, recognition of face
Module, the panel end continuous collecting face source data, and send to the face detection module, the demographic data storehouse storage
There is personnel's essential information;
When carrying out user's registration operation, the face detection module detects face characteristic in the face source data, will
It is created as master pattern, and sends to demographic data storehouse storage, is matched with personnel's essential information;
When carrying out user's identification operation, the face detection module detects face characteristic in the face source data, will
It is created as detection model, and sends to the face recognition module, and the face recognition module transfers the demographic data storehouse
The interior master pattern is matched with the detection model, obtains matching rate.
2. face identification system according to claim 1, it is characterised in that:The face identification system also includes face mould
Plate database, the face template database stores the basic model of common face;
When carrying out user's registration operation, the face detection module detects face characteristic in the face source data, and
The suitable basic model is selected in the face template database based on the face characteristic, the face characteristic is filled out
Fill in the basic model, so as to be created that the master pattern, the master pattern is sent to the demographic data storehouse and stored up
Deposit, matched with personnel's essential information;
When carrying out user's identification operation, the face detection module is similarly created that the detection model, by the detection mould
Type is sent to the face recognition module, and the face recognition module transfers the master pattern and detection model progress
Match somebody with somebody, obtain matching rate.
3. face identification system according to claim 2, it is characterised in that:The face identification system also includes tracking mould
Block;
The matching rate of the acquisition of the face recognition module is temporarily stored into the tracking module, and the tracking module drives described
Face detection module creates another detection model again, and the face recognition module is again to detection model progress
Match somebody with somebody, obtain another matching rate;
Repeat after certain number of times, the tracking module exports recognition result with reference to multiple matching rates to the panel end.
4. according to any described face identification system of claims 1 to 3, it is characterised in that:Adopted including data at the panel end
Collect unit, switch unit, display unit, the data acquisition unit continuous collecting face source data, and send to the face
Detection module;
During user's registration operation to be carried out, the face detection module is switched to by login state by the switch unit, will
The face characteristic is created as the master pattern;
During user's identification operation to be carried out, the face detection module is switched to by identification state by the switch unit, will
The face characteristic is created as the detection model;
The tracking module exports the recognition result to the display unit.
5. face identification system according to claim 1, it is characterised in that the detection method of the face identification system,
Comprise the following steps:
Step one:Pretreatment, detects the edge of portrait in the face source data, removes the high intensity in the face source data
Noise, extract shape of face feature and texture feature, be designated as foundation characteristic;
Step 2:Local Characteristic is analyzed:Set and eliminate edge effect, the induction coefficient of light environment error, be according to the sensing
Based on number, the edge effect in the foundation characteristic and light environment error are shielded, effective detection result is obtained, is designated as face
Feature;
Step 3:Dynamic Local specificity analysis:Induction coefficient described in the step 2 is changed, another face is obtained special
Levy;
Step 4:Interative computation:Preferably one facial characteristics is chosen in the step 2 and the step 3 as institute
Foundation characteristic is stated, rerun the progress step 2 and the step 3, until obtaining the suitable facial characteristics conduct
The face characteristic.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710380058.7A CN107194357A (en) | 2017-05-25 | 2017-05-25 | Face identification system and its detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710380058.7A CN107194357A (en) | 2017-05-25 | 2017-05-25 | Face identification system and its detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107194357A true CN107194357A (en) | 2017-09-22 |
Family
ID=59875561
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710380058.7A Pending CN107194357A (en) | 2017-05-25 | 2017-05-25 | Face identification system and its detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107194357A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110070037A (en) * | 2019-04-22 | 2019-07-30 | 深圳力维智联技术有限公司 | Smooth upgrading method, device and the readable storage medium storing program for executing of human face recognition model |
RU2741412C1 (en) * | 2020-03-19 | 2021-01-25 | Общество с ограниченной ответственностью «Арсенал 67» | Face recognition system for comparison with databases using mobile terminals |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106156688A (en) * | 2015-03-10 | 2016-11-23 | 上海骏聿数码科技有限公司 | A kind of dynamic human face recognition methods and system |
CN106339679A (en) * | 2016-08-24 | 2017-01-18 | 奥奈斯特(宁波)软件技术有限公司 | Face recognition system |
-
2017
- 2017-05-25 CN CN201710380058.7A patent/CN107194357A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106156688A (en) * | 2015-03-10 | 2016-11-23 | 上海骏聿数码科技有限公司 | A kind of dynamic human face recognition methods and system |
CN106339679A (en) * | 2016-08-24 | 2017-01-18 | 奥奈斯特(宁波)软件技术有限公司 | Face recognition system |
Non-Patent Citations (3)
Title |
---|
徐挺: "大容量人脸识别门禁***的研究", 《计算机仿真》 * |
王大女: "基于离散模板精匹配的人脸识别算法", 《兵工自动化》 * |
辜小花: "基于视觉感知和边缘保持的光照不变人脸识别", 《电子学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110070037A (en) * | 2019-04-22 | 2019-07-30 | 深圳力维智联技术有限公司 | Smooth upgrading method, device and the readable storage medium storing program for executing of human face recognition model |
CN110070037B (en) * | 2019-04-22 | 2022-11-01 | 深圳力维智联技术有限公司 | Smooth upgrading method and device for face recognition model and readable storage medium |
RU2741412C1 (en) * | 2020-03-19 | 2021-01-25 | Общество с ограниченной ответственностью «Арсенал 67» | Face recognition system for comparison with databases using mobile terminals |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110298291B (en) | Mask-RCNN-based cow face and cow face key point detection method | |
CN110443231A (en) | A kind of fingers of single hand point reading character recognition method and system based on artificial intelligence | |
CN101615311B (en) | Method for evaluating queuing time based on vision | |
CN105718873B (en) | Stream of people's analysis method based on binocular vision | |
CN101853399B (en) | Method for realizing blind road and pedestrian crossing real-time detection by utilizing computer vision technology | |
CN110490238A (en) | A kind of image processing method, device and storage medium | |
CN105007395A (en) | Privacy processing method for continuously recording video | |
CN104392229B (en) | Person's handwriting recognition methods based on stroke direction of fragments distribution characteristics | |
CN102222216A (en) | Identification system based on biological characteristics of fingerprints | |
CN206322194U (en) | A kind of anti-fraud face identification system based on 3-D scanning | |
CN106096662A (en) | Human motion state identification based on acceleration transducer | |
CN102567993A (en) | Fingerprint image quality evaluation method based on main component analysis | |
CN109544754A (en) | A kind of guard method and system based on recognition of face, computer equipment | |
CN104156729B (en) | A kind of classroom demographic method | |
CN101464945A (en) | Identification characteristic identification method based on finger back arthrosis veins | |
CN109993130A (en) | One kind being based on depth image dynamic sign language semantics recognition system and method | |
CN104504360A (en) | Automatic authentication method for ancient ceramics | |
CN109063643A (en) | A kind of facial expression pain degree recognition methods under the hidden conditional for facial information part | |
CN103714159B (en) | Merge two grades and three grades features by slightly to smart fingerprint identification method | |
CN107194357A (en) | Face identification system and its detection method | |
US20200293781A1 (en) | Food detection and identification method based on deep learning | |
CN111144359A (en) | Exhibit evaluation device and method and exhibit pushing method | |
CN110427943A (en) | A kind of intelligent electric meter technique for partitioning based on R-CNN | |
CN103714331A (en) | Facial expression feature extraction method based on point distribution model | |
KR100393753B1 (en) | Method for recognizing fingerprint using dual mode matching |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170922 |