CN103514447A - Face recognition system based on smart phone - Google Patents
Face recognition system based on smart phone Download PDFInfo
- Publication number
- CN103514447A CN103514447A CN201310490162.3A CN201310490162A CN103514447A CN 103514447 A CN103514447 A CN 103514447A CN 201310490162 A CN201310490162 A CN 201310490162A CN 103514447 A CN103514447 A CN 103514447A
- Authority
- CN
- China
- Prior art keywords
- image
- mobile phone
- face
- smart mobile
- face recognition
- 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
Abstract
A face recognition system based on a smart phone comprises the following face recognition processes: a user shoots facial features to be confirmed through a shooting function of the smart phone, and sends a picture to a server; the server matches and compares the picture with an image stored in a database by using an image feature extractor to find out picture information with high confirmation degree, and sends the picture information back to the mobile phone user for people confirmation. The face recognition system not only has confirming and verifying functions of a general face recognition system, makes full use of advantages of wireless communication, and has the advantages of mobility and portability, so that the face recognition system has a relatively wide application prospect.
Description
Technical field
The present invention relates to a kind of face identification system, be specifically related to a kind of based on smart mobile phone face identification system.
Background technology
Recognition of face is refered in particular to by personage's face and is carried out identity validation or identity is searched.At present, it is ripe that face recognition technology has become, and dissimilar commercial system comes into operation.Face identification system carries out video camera and automatically identifies discovery by setting up the specific people of Automatic face recognition warning net ，Dui specific region, authorizes the personnel that enter specific region without registration, and system i.e. directed warning.Current face identification system equipment volume is larger, and travelling performance is poor, is not easy to carry, and is difficult to universal and widespread use.
CBIR (Content Based Image Retrieval) is CBIR.The research field that belongs to graphical analysis and information processing. refer to that direct employing picture material carries out the inquiry of image information, object is under the prerequisite of given query image, query criteria according to its content information or appointment, in image data base, carry out consistent in content or similarity matching, the respective image that meets querying condition is finally provided.
The communication technology is maked rapid progress, smart mobile phone the function such as to shoot with video-corder further powerful and universal, the portable face identification system that designs fusion CBIR technology and mobile communication technology has become possibility.
Summary of the invention
The object of the present invention is to provide a kind of based on smart mobile phone face identification system, it not only has, and general face identification system is confirmed and the function of verification, and take full advantage of the advantage of wireless telecommunications, there is movability, portable, thereby have application prospect more widely.
Technical solution of the present invention is:
A kind of based on smart mobile phone face identification system, the face recognition process that it is characterized in that this system is as follows: the personage that user confirms needs by the camera function of smart mobile phone self carries out facial characteristics shooting, picture is sent to server, server by utilizing image characteristics extraction device carries out matching ratio to processing by the image storing in itself and database, the pictorial information that the degree of finding out is high, sent it back again cellphone subscriber, for carrying out personage's confirmation.
Above-mentioned based on smart mobile phone face identification system, it is characterized in that the communication between smart mobile phone and server is the mode communication with note or micro-letter or QQ transmission.
Above-mentioned database comprises image data base and characteristics of image storehouse, in described image data base, to extract and the ordering treatment such as index facial characteristics, and memory image, the face line coding of described characteristics of image library storage image facial characteristics.
Above-mentioned image characteristics extraction device comprises that proper vector method and face line template are two kinds of main algorithm that extraction apparatus is used; Described proper vector method is first to determine the attributes such as the size of the image surface face profiles such as an iris, the wing of nose, the corners of the mouth, position, distance, and then calculates their geometric feature, and these characteristic quantities form one and describe this facial proper vector; In described line template Shi storehouse, store some index planes as template or image surface organ template, when comparing, adopt normalization correlative tolerance to mate all pixels of sampling image surface and all templates in storehouse.
The invention has the advantages that: successfully merge and applied the existing technological achievements such as CBIR technology, Internet technology and mobile communication, utilized the advantage of wireless telecommunications, make smart mobile phone possess the function that face identification system is confirmed and investigated, there is movability, portable, thereby have application prospect more widely.
Embodiment
CBIR generally consists of image indexing system and image indexing system two parts.Image indexing system thumbnail file also arranges retrieval heading by designing requirement and retrieves a little, and forming one can be for the orderly heading directory system of coupling retrieval.This system provides the retrieval population of the different images bottom visual signatures such as color, texture, shape and object by design function.The main retrieval of content of CBIR has color, texture, shape and object etc.Color characteristic comprises color of image distribution, mutual relationship and composition etc.; Texture refers to image texture structure, direction, combination and symmetric relation etc.; Shape refers to image outline composition, shape, size etc.; Object comprises relation, quantity, attribute and the rotation etc. of image subobject.
CBIR can directly extract feature and semanteme from image, and retrieving is directly connected with extraction of semantics, makes retrieving more effective, and adaptability is stronger; By similar coupling (Similar Match), replace exact matching (Exact Match), right method obtains similar picture structure to adopt ratio of similitude, asymptotic convergent, until obtain satisfactory result; User can select example or own graphing to inquire about by browsing, and can update retrieval type, refinement retrieving; 205 retrievals based on objective attribute (keyword) are provided, and content-based retrieval, based on multi-level efficient retrieval such as object association retrieval and conceptual retrievals.
In CBIR field, through the theoretical research of more than ten years, algorithm and some valuable systems of many comparative maturities have been produced.Now generally adopt the image information (as color of image, texture, shape etc.) of low level to realize query by image content.QBIC (the QueryBy Image Content) image indexing system of RuIBM research centre exploitation, the Visual SEEK image query systems of Columbia University, the PhotoBook system of Massachusetts Institute of Technology's development in laboratory, [6] such as MARS systems of U.S. UIUC university.For further improving the accuracy of retrieval, CBIR system adopts similarity algorithm, calculate user and submit the similarity size recording in result and index data base to, extracting the information that meets threshold value as a result of and according to the mode of similarity descending exports, and constantly carrying out alternately with user with in a retrieving, system is learnt by the relevant information to these feedbacks, again carries out next round retrieval, thereby reaches user's requirement.
Web application example based on CBIR technology
System in above-mentioned " application system based on CBIR technology " mainly utilizes color, texture, shape of image etc. to carry out similarity comparison, majority is that the bottom visual signature based on image is retrieved, there is no complete Web implementation, distance is generally used greatly differ from each other [7].At present, complete CBIR technology application website of realizing Webization mainly contains www.1ike.COrnwww.polarrose.corn.Www.1ike.corn is a gyp dress ornament shopping website, and it has realized the retrieval and indexing to dress ornament image with unique technology.User submits the photo of clothes, shoes and hats, jewelry etc. to, retrieves Hou， website can return to the relevant informations such as the name of an article of this dress ornament, price by coupling.The major limitation of this system is that the content of search can only be dress ornament, and is world-famous product.What is interesting is and submit famous person's photo to, just can search out the thing that famous person dresses.Return to the price list of a series of similar dress ornaments for shopping reference.Www.polarrose.tom utilizes its own two dimensional image three-dimensional model transformation technology, and free face-image retrieval service is provided.This service realizes in conjunction with the operating software of client and the processing capacity of server end.Client software is with the client software form external disclosure of computer network browser Firefox and IE.When user browses web sites and show the still image that contains personage's face-image, above-mentioned client software will produce 21 little signs at personage's face, clicks this sign, just can retrieve face and image like this figure kind.If this personage's name has registered to PolarRose database, also can confirm its name.If the personage who selects is User ID Assigned not yet, user is User ID Assigned or right the wrong voluntarily also
Name.Vice president's Nicholas of PolarRose company is described its function like this: " utilize our people's face search technique; you just can be according to some essential characteristics of people face; realization is sought 206 and ' looked for same person, as long as can see his face clearly in the photo of different scenes, light.”
The present invention is based on the reaching its maturity and the Successful Practice of Web implementation of universal, CBIR technology of Internet, CBIR, Internet and the Trinitarian design philosophy of wireless communication technique are proposed, to realize a kind of novel Mobile face identification system that mobile phone is carrier of take.
Principle of work and the face recognition process of this system are as follows: the personage that user confirms needs by the camera function of smart mobile phone self carries out facial characteristics shooting, picture is sent to server with SMS form, server carries out the image storing in itself and database after " content-based " matching treatment, again the high pictorial information of matching degree is sent it back to cellphone subscriber with note form, for carrying out personage's confirmation.
System operational scheme
Personage's face image data storehouse of model CBIR technical support will be extracted and the ordering treatment such as index facial characteristics in personage's face image data storehouse, and then the facial characteristics storehouse of the storage face of foundation line coding.This is basis and the prerequisite of system works.Under the support of this basis and prerequisite, the image that user takes mobile phone is sent to server by wireless network, the image receiving is carried out to facial feature extraction, then in feature database, mate retrieval, the pictorial information that similarity is high postbacks to cellphone subscriber by SMS form, for identity validation.
System module explanation
(1) database
Image data base: the character image and the comment thereof that utilize Spider Web Spider program to capture in webpage deposit picture database in, in order to make up the quantitative deficiency of professional image library; Professional image and annotation thereof, consist of, object is to provide authoritative explanation for user.
Characteristics of image storehouse: extract the image facial characteristics in image data base 1, deposit face line coding in characteristics of image storehouse 1; Be used for depositing the face line coding of image in image data base 2." face line coding " is to carry out work according to the essential characteristic of face and shape. it can resist the variation of light, skin color, facial hair, hair style, glasses, expression and attitude, there is powerful reliability, can from million people, accurately recognize a people.
(2) image characteristics extraction device
This module is comprised of algorithm, and proper vector method and face line template are two kinds of main algorithm that extraction apparatus is used.Proper vector method is first to determine the attributes such as the size of the image surface face profiles such as an iris, the wing of nose, the corners of the mouth, position, distance, and then calculates their geometric feature, and these characteristic quantities form one and describe this facial proper vector; In face line template Shi storehouse, store some index planes as template or image surface organ template, when comparing, adopt normalization correlative tolerance to mate all pixels of sampling image surface and all templates in storehouse.In addition, adopt in addition the auto-correlation network of pattern-recognition or the method that feature combines with template.
(3) index
Index is a whole order module of image, and it carries out category index to image data base and characteristics of image storehouse.It is optimization data storehouse institutional framework and the effective ways that improve system works efficiency that image data base is carried out to index, and in fact it is that the face line coding of image is carried out to ordering treatment, can be effective
Dwindle range of search, improve the response speed of system.
(4)GUI
GUI (Graphical User Interface) is visual experience and the interactive operation part of screen product.GUI is the man-machine system engineering of a kind of combination computer science, aesthetics, psychology, behaviouristics and each commercial field demand analysis, emphasizes that people one machine one environment three carries out overall design as a system.The object of this customer-oriented System Engineering Design is the performance of optimizing product, makes to operate more humane, alleviates user's cognitive load, makes its operational requirements that is more suitable for user, and the GUI of native system relates generally to network inquiry part and mobile phone part.
(5) main interface
Mobile phone is connected by base station and system server, mainly by CMPP or SGIP agreement, realizes.CMPP and SGIP (being ETIP on CD-MA) agreement is respectively the short message internet access scheme that solution China Mobile and CHINAUNICOM provide, the Applied layer interface agreement of their provisioning information station resource entities and internet short message gateway.CMPP and SGIP can provide service for realizing mobile data value-added services, comprise following business: Email notice, voicemail notification, Internet send short message, mobile platform is sent out Email, urged expense notice, automatic Synthesis business information platform.
Claims (4)
1. one kind based on smart mobile phone face identification system, the face recognition process that it is characterized in that this system is as follows: the personage that user confirms needs by the camera function of smart mobile phone self carries out facial characteristics shooting, picture is sent to server, server by utilizing image characteristics extraction device carries out matching ratio to processing by the image storing in itself and database, the pictorial information that the degree of finding out is high, sent it back again cellphone subscriber, for carrying out personage's confirmation.
2. according to claim 1 based on smart mobile phone face identification system, it is characterized in that the communication between smart mobile phone and server is the mode communication with note or micro-letter or QQ transmission.
3. according to claim 1 based on smart mobile phone face identification system, it is characterized in that described database comprises image data base and characteristics of image storehouse, in described image data base, to extract and the ordering treatment such as index facial characteristics, and memory image, the face line coding of described characteristics of image library storage image facial characteristics.
4. according to claim 1 based on smart mobile phone face identification system, it is characterized in that described image characteristics extraction device comprises that proper vector method and face line template are two kinds of main algorithm that extraction apparatus is used; Described proper vector method is first to determine the attributes such as the size of the image surface face profiles such as an iris, the wing of nose, the corners of the mouth, position, distance, and then calculates their geometric feature, and these characteristic quantities form one and describe this facial proper vector; In described line template Shi storehouse, store some index planes as template or image surface organ template, when comparing, adopt normalization correlative tolerance to mate all pixels of sampling image surface and all templates in storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310490162.3A CN103514447A (en) | 2013-10-18 | 2013-10-18 | Face recognition system based on smart phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310490162.3A CN103514447A (en) | 2013-10-18 | 2013-10-18 | Face recognition system based on smart phone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103514447A true CN103514447A (en) | 2014-01-15 |
Family
ID=49897141
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310490162.3A Pending CN103514447A (en) | 2013-10-18 | 2013-10-18 | Face recognition system based on smart phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103514447A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104143086A (en) * | 2014-07-18 | 2014-11-12 | 吴建忠 | Application technology of portrait comparison to mobile terminal operating system |
| CN104317827A (en) * | 2014-10-09 | 2015-01-28 | 深圳码隆科技有限公司 | Picture navigation method of commodity |
| CN104679832A (en) * | 2015-02-05 | 2015-06-03 | 四川长虹电器股份有限公司 | System and method for searching single or multi-body combined picture based on face recognition |
| CN106953408A (en) * | 2017-04-19 | 2017-07-14 | 湖南美奕机电科技有限公司 | A kind of monitoring system of electric substation |
| CN107992795A (en) * | 2017-10-27 | 2018-05-04 | 江西高创保安服务技术有限公司 | Clique and its head's recognition methods based on people information storehouse and real name message registration |
| CN108156276A (en) * | 2017-12-28 | 2018-06-12 | 华慧视科技（天津）有限公司 | A kind of circle of friends management method based on recognition of face |
| CN116012924A (en) * | 2023-01-30 | 2023-04-25 | 人民网股份有限公司 | Face gallery construction method and device and computing equipment |
-
2013
- 2013-10-18 CN CN201310490162.3A patent/CN103514447A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104143086A (en) * | 2014-07-18 | 2014-11-12 | 吴建忠 | Application technology of portrait comparison to mobile terminal operating system |
| CN104317827A (en) * | 2014-10-09 | 2015-01-28 | 深圳码隆科技有限公司 | Picture navigation method of commodity |
| CN104317827B (en) * | 2014-10-09 | 2017-06-30 | 深圳码隆科技有限公司 | A kind of picture air navigation aid of commodity |
| CN104679832A (en) * | 2015-02-05 | 2015-06-03 | 四川长虹电器股份有限公司 | System and method for searching single or multi-body combined picture based on face recognition |
| CN106953408A (en) * | 2017-04-19 | 2017-07-14 | 湖南美奕机电科技有限公司 | A kind of monitoring system of electric substation |
| CN107992795A (en) * | 2017-10-27 | 2018-05-04 | 江西高创保安服务技术有限公司 | Clique and its head's recognition methods based on people information storehouse and real name message registration |
| CN107992795B (en) * | 2017-10-27 | 2021-08-31 | 江西高创保安服务技术有限公司 | Group partner based on population information base and real name call record and its head and target identification method |
| CN108156276A (en) * | 2017-12-28 | 2018-06-12 | 华慧视科技（天津）有限公司 | A kind of circle of friends management method based on recognition of face |
| CN116012924A (en) * | 2023-01-30 | 2023-04-25 | 人民网股份有限公司 | Face gallery construction method and device and computing equipment |
| CN116012924B (en) * | 2023-01-30 | 2023-06-27 | 人民网股份有限公司 | Face gallery construction method and device and computing equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103514447A (en) | Face recognition system based on smart phone | |
| CN104239408B (en) | The data access of content based on the image recorded by mobile device | |
| CN100578508C (en) | Interactive type image search system and method | |
| US11501514B2 (en) | Universal object recognition | |
| AU2016291660B2 (en) | Apparatus and methods for facial recognition and video analytics to identify individuals in contextual video streams | |
| WO2017167060A1 (en) | Information display method, device and system | |
| CN102567543B (en) | Clothing picture search method and clothing picture search device | |
| CN106933867B (en) | Image query method and device | |
| CN108369652A (en) | The method and apparatus that erroneous judgement in being applied for face recognition minimizes | |
| CN103336796A (en) | Method and system for displaying door buster directly | |
| CN105824928A (en) | Mobile terminal, server, content-based image recognition searching method and system | |
| CN104036002A (en) | Technical method for intelligently recommending data | |
| CN105677728A (en) | Object image recognition and classification managing method | |
| CN101308571A (en) | Method for generating novel human face by combining active grid and human face recognition | |
| CN103412938A (en) | Commodity price comparing method based on picture interactive type multiple-target extraction | |
| WO2016202132A1 (en) | Method and system for photo management | |
| CN102880634A (en) | Intelligent face recognition and retrieval method based on cloud | |
| CN103428537A (en) | Video processing method and video processing device | |
| CN105868686A (en) | Video classification method and apparatus | |
| CN106959993A (en) | The position tracking method of reserve | |
| KR20210002410A (en) | System, method and program of constructing dataset for training appearance recognition model | |
| CN105117399A (en) | Image search method and device | |
| CN105095333A (en) | Mobile internet based lost and found method and system | |
| TW201719518A (en) | Searching method for product photograph and searching system for product photograph | |
| CN103258190A (en) | Face recognition method used for mobile terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1193488 Country of ref document: HK |
|
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140115 |
|
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: WD Ref document number: 1193488 Country of ref document: HK |