CN101021899A - Interactive human face identificiating system and method of comprehensive utilizing human face and humanbody auxiliary information - Google Patents
Interactive human face identificiating system and method of comprehensive utilizing human face and humanbody auxiliary information Download PDFInfo
- Publication number
- CN101021899A CN101021899A CN 200710020642 CN200710020642A CN101021899A CN 101021899 A CN101021899 A CN 101021899A CN 200710020642 CN200710020642 CN 200710020642 CN 200710020642 A CN200710020642 A CN 200710020642A CN 101021899 A CN101021899 A CN 101021899A
- Authority
- CN
- China
- Prior art keywords
- face
- module
- information
- human body
- supplementary
- 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.)
- Granted
Links
Images
Landscapes
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
This invention relates to an interaction man-face identification system and a method for utilizing man-face and human body assistant information including a customer end, a server and a communication line between them, in which, the customer end includes a man-face detection module, a character pick-up module and a user confirmation module, the server includes a man-face database module, an engine identifying module and the adjustment of the engine-identifying module or the server includes a network man-face database module, the customer end includes local man-face database module, a man-face test module, a character pick-up module, an engine-identifying module and a user confirmation module used in confirming the identified result and adjusting identification engines based on the confirmed result.
Description
Technical field
The present invention relates to image identification system and method, specifically the interactive face identification system and the method for comprehensive utilization personnel selection face and human body assistant information.
Background technology
Existing face identification system is mainly used in occasions such as security monitoring, the system of sealing on a small scale normally, and in these systems, mostly just single utilization people's face information discern.The patent No. is: 200310120624.9, and patent name is: the Chinese patent of " distributed people's face detects and recognition methods under the mobile computing environment " is a face identification system that is applicable to the mobile network based on user terminal/server framework.But this system has only utilized people's face information.People's face information is subjected to the influence of external environment condition (such as illumination) and human face posture easily, only utilizes people's face information, and recognition performance has certain limitation, and simultaneously, this system does not introduce user's feedback, thereby can't utilize user's feedback to improve the performance of identification.Chinese patent application number is: 02153265.6, patent name is: the Chinese patent of " utilizing body information matched human face information's personal identification method ", its ultimate principle is at first to come human body by motion detection, detect people's face again, utilize body information and people's face information to discern then.There are several shortcomings in this system: at first, because need motion detection, this method is not suitable for rest image; Secondly, body information need collect complete human body, and this condition does not satisfy in many utility systems; Once more, the accuracy rate of human detection is not high usually, thereby the accuracy rate of total system is not high; Once more, this system does not introduce user's feedback, thereby can't utilize user's feedback to improve the performance of identification; Once more, this system does not clearly introduce the notion of time, and the weighting of height information and people's face information fixes in advance, but can know that many times height information is inaccurate, and the weighting meeting of machinery causes identification error; At last, this system is not the system of a networking.
Summary of the invention
The objective of the invention is to propose a kind of precision that improves identification, can be applied to the large-scale based on the comprehensive utilization personnel selection face of client, server architecture and the interactive face identification system and the method for human body assistant information of internet, the present invention can be widely used in static and the live image.
Purpose of the present invention can be achieved through the following technical solutions:
The interactive face identification system of comprehensive utilization personnel selection face and human body assistant information comprises client, server, and communication line between the two,
Client comprises with lower module:
People's face detection module to the image to be identified of input, detects people's face wherein;
Characteristic extracting module for detected people's face, is extracted the supplementary of people's face and human body, and the photo opporunity of picture, the camera type information, and the correlated characteristic vector reached server by communication line;
The user confirms module, is used to confirm that server passes the recognition result of coming, and the recognition result of confirming is reached server by communication line;
Server comprises with lower module:
The face database module provides the personage's who has trained proper vector;
The recognition engine module, the personage's that acquisition has been trained from face database proper vector, and the proper vector that passes the personage to be identified come from client discerned in contrast, obtain recognition result, and recognition result is reached the user by communication line confirm module;
Adjust the recognition engine module, confirm that according to the user recognition result of the affirmation that the module biography is come is adjusted recognition engine.
The interactive face identification system of comprehensive utilization personnel selection face of the present invention and human body assistant information can also be formed structure for another kind, comprises client, server, and communication line between the two,
Server comprises with lower module:
Netter's face database module provides the character features vector data;
Client comprises with lower module:
Native's face database module before beginning identification, according to the task of identification, to the character features vector set of module request of server network face database and identification mission coupling, and downloads in the local face database;
People's face detection module to the image to be identified of input, detects people's face wherein;
Characteristic extracting module for detected people's face, is extracted the supplementary of people's face and human body, and the photo opporunity of picture, the camera type information;
The recognition engine module obtains the proper vector of the people's face trained from local figure database, and personage's to be identified proper vector is discerned in contrast, obtains recognition result;
The user confirms module, is used to confirm recognition result;
Adjust the recognition engine module, adjust recognition engine according to the recognition result of confirming.
Realize that the comprehensive utilization personnel selection face of the object of the invention and the interactive face identification method of human body assistant information may further comprise the steps:
(1) detects people's face automatically;
(2) determine the characteristics of human body zone according to the detected people's face of step (1);
(3) photo opporunity, the camera type information of extraction people's face and human body assistant information and picture;
(4) get rid of non-special human body assistant information;
(5) comprehensive utilization personnel selection face and special supplementary and photo opporunity, the camera type information of picture are discerned, and obtain recognition result.
Purpose of the present invention can also further realize by following technical measures:
The interactive face identification system of aforesaid comprehensive utilization personnel selection face and human body assistant information, characteristic extracting module comprises following submodule: human face characteristic point locator module, for detected people's face, location human face characteristic point wherein; Facial image conversion process submodule according to detected unique point, carries out normalization and affined transformation to facial image, makes two to be positioned at the fixed position, and the facial image of intercepting fixed size; The standardization processing submodule adopts image is carried out histogram equalization or normalization or to carrying out Fuzzy processing or according to attitude image is carried out based on faceform's ways of geometric illumination, background, attitude being carried out standardization processing in the image border; Face characteristic vector extracts submodule, adopts principal component analysis (PCA) or the method that image characteristic point extracts wavelet conversion coefficient is extracted face characteristic vector in the facial image; Human body assistant information is extracted submodule, extracts clothes information, hair information, picture photo opporunity, camera type information.
The interactive face identification system of aforesaid comprehensive utilization personnel selection face and human body assistant information, the recognition engine module comprises the contrast submodule: calculate the distance between two proper vectors, Euclidean distance or X
2Distance, if personage's supplementary is arranged, judging that this supplementary is under the unique situation, and shooting time is approaching, then people's face distance and the distance weighted addition of supplementary, otherwise end user's face distance only, in a plurality of candidate feature vectors, adopt nearest neighbor method, i.e. the known features vector of selected distance minimum, if there is not special supplementary, the face information of then only choosing.
The interactive face identification system of aforesaid comprehensive utilization personnel selection face and human body assistant information, adjust the recognition engine module and comprise at least with next sub-module: correct recognition result warehouse-in submodule is added to face database to the result of correct identification; Wrong identification result puts submodule in storage, and the result of identification error is recorded face database with non-someone form, repels this result when being provided with back identification; Recognition engine is adjusted submodule, according to the result of identification feedback, adopts the method for relevant feedback to adjust recognition engine.
The interactive face identification method of aforesaid comprehensive utilization personnel selection face and human body assistant information, in the step (1), the method that detects people's face is the Adaboost method.
The interactive face identification method of aforesaid comprehensive utilization personnel selection face and human body assistant information, in the step (2), according to size, the position of detected people's face, rule of thumb value is determined the characteristics of human body zone.For example, rule of thumb value is set up a masterplate, uses this masterplate according to size, the position of detected people's face and determines the characteristics of human body zone.
The interactive face identification method of aforesaid comprehensive utilization personnel selection face and human body assistant information, in the step (3), extracting human body assistant information is to adopt the method for color histogram at characteristics of human body's extracted region colouring information, employing is carried out frequency domain transform in the characteristics of human body zone, the method of extracting frequency domain distribution information is at characteristics of human body's extracted region texture information, also can extract local binary pattern (LBP, the Locally Binary Pattern) information of image.
The interactive face identification method of aforesaid comprehensive utilization personnel selection face and human body assistant information, step (5) may further comprise the steps: it is identical 1. at first to choose the camera model, the picture that photo opporunity is close; 2. by the supplementary in the scanning group photo, if the distance of two people's supplementary is less than certain threshold value in the group photo, then think non-special supplementary, perhaps for people's face of the mark of user, the supplementary that compares them in twos, if the distance of non-same individual and two supplementarys is less than certain threshold value, then think non-special supplementary, non-special supplementary is stored in the database, non-special supplementary contrast in supplementary to be identified and the database, if less than certain threshold value, then determine not utilize this supplementary to discern this personage; 3. comprehensive utilization personnel selection face information and special supplementary are discerned: the distance of weighting people face information and special supplementary, and weighted information is recorded in database, if there is not special supplementary, the face information of then only choosing; 4. the user confirms recognition result; 5. according to confirming that the result adjusts recognition engine, if last time recognition result had utilized a large amount of supplementarys and the recognition result mistake then joins this supplementary in the non-special side information data storehouse.
Advantage of the present invention is: the present invention is the face identification system of a kind of comprehensive utilization personnel selection face and human body assistant information, comprehensive utilization personnel selection face and human body assistant information can improve the precision of identification, can discern the impalpable personage of conventional method by human body assistant information.The present invention is a kind of interactively face identification system, by introducing user feedback and adjusting recognition engine, can improve the precision of identification.In addition, the present invention can be applied to the internet, is large-scale face identification system based on client, server architecture, has solved the concurrent request of supporting a large number of users from framework.
Description of drawings
Fig. 1 is the system chart of the embodiment of the invention one.
Fig. 2 is the system chart of the embodiment of the invention two.
Embodiment
Embodiment one
The system chart of present embodiment as shown in Figure 1, the interactive face identification system of a kind of comprehensive utilization personnel selection face and human body assistant information comprises client, server, and communication line between the two.Client can be personal computer (PC), also can be the hardware terminal (based on special chip or digital signal processor etc.) of special use, also can be portable terminal (as mobile phone).Communication line can be the internet, also can be wireless network, can also be dedicated communication line the dedicated communication line of fields such as security protection (as be used for).Client comprises with lower module: people's face detection module, characteristic extracting module, user confirm module, and server comprises with lower module: face database module, recognition engine module, adjustment recognition engine module.
People's face detection module detects people's face wherein to the image to be identified of input, and people's face detects the method that can adopt Adaboost.The Adaboost method is at Paul Viola and MichaelJones, Rapid object detection using a boosted cascade of simplefeatures, and CVPR has a detailed description in 2,001 one literary compositions, does not repeat them here.
Characteristic extracting module is carried out the proper vector extraction and be may further comprise the steps: for detected people's face, and location human face characteristic point wherein; According to detected unique point, facial image is carried out normalization and affined transformation, make two to be positioned at the fixed position, the facial image of intercepting fixed size; Illumination, background, attitude are carried out standardization processing, and the method that can adopt comprises carries out histogram equalization or normalization to image, and Fuzzy processing is carried out in the image border, image is carried out geometric transformation based on the faceform etc. according to attitude.On this basis, extract the face characteristic in the facial image, the method that can adopt is principal component analysis (PCA) (PCA), or image characteristic point is extracted wavelet transformation (as the Gabor conversion) coefficient, obtains the proper vector of people's face to be identified thus.Except people's face information, can also extract the supplementary of human body, can be used as useful identifying information, comprise hair, clothes, the bodily form etc.In addition, the photo opporunity of picture, the camera model also is a Useful Information, can pass to server simultaneously.People's face information, the supplementary of human body and the photo opporunity of picture, the camera model is together to form a personage's proper vector.The extraction of the information of human body, we are example with the clothes, specifically carry out as follows: selected clothes zone, this can be according to size, the position, selected according to empirical value of detected people's face; At clothes extracted region colouring information, this can adopt color histogram; At clothes extracted region texture information, this can adopt and carry out frequency domain transform (as small echo, GABOR, the discrete surplus conversion etc. of revolving) in the clothes zone, extract the distributed intelligence of frequency domain, perhaps extract local binary pattern (LBP, the Locally Binary Pattern) information of image; Extract photo opporunity, the camera type information of picture.
Client is passed to server to personage's to be identified proper vector by communication line, and the recognition engine module in the server obtains the personage's that trained proper vector from face database, and personage's to be identified proper vector is discerned in contrast.Two distances between the proper vector are normally calculated in comparison, and this distance can adopt Euclidean distance, X
2Distance etc.If personage's supplementary is arranged, judging that this supplementary is under the unique situation, and shooting time is approaching, then people's face distance and the distance weighted addition of supplementary, if there is not special supplementary, the face information of then only choosing.In a plurality of candidate feature vectors, can adopt nearest neighbor method (kNN), i.e. the known features vector of selected distance minimum, if this distance less than some predetermined threshold values, the name of this proper vector is personage's to be identified name.
Server is passed to client to the result of identification by communication line, the user confirms recognition result, client is passed to server to the affirmation of recognition result, adjusts the recognition engine module in the server and carries out one of following or all: the result who correctly discerns is added to face database; The result of identification error is recorded face database with non-someone form, repel this result when being provided with back identification; According to the result of identification feedback, adjust recognition engine with the method for relevant feedback (Relevance Feedback).A kind of specific implementation method of relevant feedback is to adjust the weight of each dimensional feature vector weighting in the recognition engine, and more weight is composed to the proper vector dimension that helps to distinguish different people.
In identification, the method for comprehensive utilization personnel selection face and supplementary comprises: it is identical at first to choose the camera model, the picture that photo opporunity is close; Determine non-special supplementary: by the supplementary in the scanning group photo, if two supplementarys are then thought non-special supplementary less than certain threshold value.The method of the non-special supplementary of another kind of decision is, for people's face of user's mark, their supplementary relatively in twos is if the distance of non-same individual and two supplementarys is then thought non-special supplementary less than certain threshold value; Non-special supplementary is stored in the database, and non-special supplementary compares in supplementary to be identified and the database, if less than certain threshold value, then determines not utilize special supplementary to discern this personage; Comprehensive utilization personnel selection face information and special supplementary are discerned, and comprehensive method is the distance of weighting people face information and special supplementary, and weighted information is recorded in database; If there is not special supplementary, the face information of then only choosing; The user confirms recognition result; According to confirming that the result adjusts recognition engine, if last time recognition result had utilized a large amount of supplementarys and the recognition result mistake then joins this supplementary in the non-special side information data storehouse.
Embodiment two
The system chart of present embodiment as shown in Figure 2, the interactive face identification system of a kind of comprehensive utilization personnel selection face and human body assistant information comprises client, server, and communication line between the two.Client can be personal computer (PC), also can be the hardware terminal (based on special chip or digital signal processor etc.) of special use, also can be portable terminal (as mobile phone).Communication line can be the internet, also can be wireless network, can also be dedicated communication line the dedicated communication line of fields such as security protection (as be used for).Client comprises with lower module: native's face database module, people's face detection module, characteristic extracting module, recognition engine module, user confirm module, and server comprises netter's face database module.
Before beginning identification, client needs the task according to identification, and the face characteristic vector set that mates to module request of server network face database and identification mission also downloads in the client terminal local face database module.
People's face detection module detects people's face wherein to the image to be identified of input, and people's face detects the method that can adopt Adaboost.The Adaboost method is at Paul Viola and MichaelJones, Rapid object detection using a boosted cascade of simplefeatures, and CVPR has a detailed description in 2,001 one literary compositions, does not repeat them here.
Characteristic extracting module is carried out the proper vector extraction, and concrete execution is identical with embodiment one.Client is defeated by the recognition engine module to personage's to be identified proper vector and supplementary.The recognition engine module obtains the proper vector of people's face of having trained from local figure database, and personage's to be identified proper vector is discerned in contrast.Two distances between the proper vector are normally calculated in comparison, and this distance can adopt Euclidean distance, X
2Distance etc.If personage's supplementary is arranged, judging that this supplementary is under the unique situation, and shooting time is approaching, then people's face distance and the distance weighted addition of supplementary, if there is not special supplementary, the face information of then only choosing.In a plurality of candidate feature vectors, can adopt nearest neighbor method (kNN), i.e. the known features vector of selected distance minimum, if this distance less than some predetermined threshold values, the name of this proper vector is personage's to be identified name.
The user confirms recognition result, according to confirming that the result adjusts recognition engine, can adopt one of following or all: the result who correctly discerns is added to face database; The result of identification error is recorded face database with non-someone form, repel this result when being provided with back identification; According to the result of identification feedback, adopt the method for relevant feedback (Relevance Feedback) to adjust recognition engine.A kind of specific implementation method of relevant feedback is to adjust the weight of each dimensional feature vector weighting in the recognition engine, and more weight is composed to the proper vector dimension that helps to distinguish different people.
The comprehensive utilization personnel selection face of present embodiment and the method for supplementary are identical with embodiment one, do not repeat them here.
The present invention can also have other embodiment, and the technical scheme that equal replacement of all employings or equivalent transformation form all drops within the scope of protection of present invention.
Claims (9)
1. the interactive face identification system of comprehensive utilization personnel selection face and human body assistant information comprises client, server, and communication line between the two, it is characterized in that:
Described client comprises with lower module:
People's face detection module to the image to be identified of input, detects people's face wherein;
Characteristic extracting module for detected people's face, is extracted the supplementary of people's face and human body, and the photo opporunity of picture, the camera type information, and the correlated characteristic vector reached server by communication line;
The user confirms module, is used to confirm that server passes the recognition result of coming, and the recognition result of confirming is reached server by communication line;
Described server comprises with lower module:
The face database module provides the personage's who has trained proper vector;
The recognition engine module, the personage's that acquisition has been trained from face database proper vector, and the proper vector that passes the personage to be identified come from client discerned in contrast, obtain recognition result, and recognition result is reached the user by communication line confirm module;
Adjust the recognition engine module, confirm that according to the user recognition result of the affirmation that the module biography is come is adjusted recognition engine.
2. the interactive face identification system of comprehensive utilization personnel selection face and human body assistant information comprises client, server, and communication line between the two, it is characterized in that:
Described server comprises with lower module:
Netter's face database module provides the character features vector data;
Described client comprises with lower module:
Native's face database module before beginning identification, according to the task of identification, to the character features vector set of module request of server network face database and identification mission coupling, and downloads in the local face database;
People's face detection module to the image to be identified of input, detects people's face wherein;
Characteristic extracting module for detected people's face, is extracted the supplementary of people's face and human body, and the photo opporunity of picture, the camera type information;
The recognition engine module obtains the proper vector of the people's face trained from local figure database, and personage's to be identified proper vector is discerned in contrast, obtains recognition result;
The user confirms module, is used to confirm recognition result;
Adjust the recognition engine module, adjust recognition engine according to the recognition result of confirming.
3. the interactive face identification system of comprehensive utilization personnel selection face as claimed in claim 1 or 2 and human body assistant information, it is characterized in that: described characteristic extracting module comprises following submodule:
Human face characteristic point locator module, for detected people's face, location human face characteristic point wherein;
Facial image conversion process submodule according to detected unique point, carries out normalization and affined transformation to facial image, makes two to be positioned at the fixed position, and the facial image of intercepting fixed size;
The standardization processing submodule adopts image is carried out histogram equalization or normalization or to carrying out Fuzzy processing or according to attitude image is carried out based on faceform's ways of geometric illumination, background, attitude being carried out standardization processing in the image border;
Face characteristic vector extracts submodule, adopts principal component analysis (PCA) or the method that image characteristic point extracts wavelet conversion coefficient is extracted face characteristic vector in the facial image;
Human body assistant information is extracted submodule, extracts clothes information, hair information, picture photo opporunity, camera type information.
4. the interactive face identification system of comprehensive utilization personnel selection face as claimed in claim 1 or 2 and human body assistant information, it is characterized in that: described recognition engine module comprises following submodule:
The contrast submodule calculates the distance between two proper vectors, Euclidean distance or X
2Distance, if personage's supplementary is arranged, judging that this supplementary is under the unique situation, and shooting time is approaching, then people's face distance and the distance weighted addition of supplementary, otherwise end user's face distance only, in a plurality of candidate feature vectors, adopt nearest neighbor method, i.e. the known features vector of selected distance minimum, if there is not special supplementary, the face information of then only choosing.
5. the interactive face identification system of comprehensive utilization personnel selection face as claimed in claim 1 or 2 and human body assistant information is characterized in that: adjust the recognition engine module and comprise at least with next sub-module:
Correct recognition result warehouse-in submodule is added to face database to the result of correct identification;
Wrong identification result puts submodule in storage, and the result of identification error is recorded face database with non-someone form, repels this result when being provided with back identification;
Recognition engine is adjusted submodule, according to the result of identification feedback, adopts the method for relevant feedback to adjust recognition engine.
6. the interactive face identification method of comprehensive utilization personnel selection face and human body assistant information is characterized in that: may further comprise the steps:
(1) detects people's face automatically;
(2) determine the characteristics of human body zone according to the detected people's face of step (1);
(3) photo opporunity, the camera type information of extraction people's face and human body assistant information and picture;
(4) get rid of non-special human body assistant information;
(5) comprehensive utilization personnel selection face and special supplementary and photo opporunity, the camera type information of picture are discerned, and obtain recognition result.
7. the interactive face identification method of comprehensive utilization personnel selection face as claimed in claim 6 and human body assistant information is characterized in that: in the described step (1), the method that detects people's face is the Adaboost method.
8. the interactive face identification method of comprehensive utilization personnel selection face as claimed in claim 6 and human body assistant information is characterized in that: in the described step (2), according to size, the position of detected people's face, rule of thumb value is determined the characteristics of human body zone.
9. the interactive face identification method of comprehensive utilization personnel selection face as claimed in claim 6 and human body assistant information, it is characterized in that: in the described step (3), extracting human body assistant information is to adopt the method for color histogram at characteristics of human body's extracted region colouring information, employing is carried out frequency domain transform in the characteristics of human body zone, the method of extracting frequency domain distribution information is at characteristics of human body's extracted region texture information, or the local binary pattern information of extraction image is as texture information.
10. the interactive face identification method of comprehensive utilization personnel selection face as claimed in claim 6 and human body assistant information, it is characterized in that: described step (5) may further comprise the steps:
1. it is identical at first to choose the camera model, the picture that photo opporunity is close;
2. determine non-special supplementary: by the supplementary in the scanning group photo, if the distance of two supplementarys is less than certain threshold value, then think non-special supplementary, perhaps for people's face of the mark of user, the supplementary that compares them in twos, if the distance of non-same individual and two supplementarys is less than certain threshold value, then think non-special supplementary, non-special supplementary is stored in the database, non-special supplementary contrast in supplementary to be identified and the database, if less than certain threshold value, then determine not utilize this supplementary to discern this personage;
3. comprehensive utilization personnel selection face information and special supplementary are discerned: the distance of weighting people face information and special supplementary, and weighted information is recorded in database, if there is not special supplementary, the distance of the face information of then only choosing;
4. the user confirms recognition result;
5. according to confirming that the result adjusts recognition engine, if last time recognition result had utilized a large amount of supplementarys and the recognition result mistake then joins this supplementary in the non-special side information data storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710020642A CN100580691C (en) | 2007-03-16 | 2007-03-16 | Interactive human face identificiating system and method of comprehensive utilizing human face and humanbody auxiliary information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710020642A CN100580691C (en) | 2007-03-16 | 2007-03-16 | Interactive human face identificiating system and method of comprehensive utilizing human face and humanbody auxiliary information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101021899A true CN101021899A (en) | 2007-08-22 |
| CN100580691C CN100580691C (en) | 2010-01-13 |
Family
ID=38709658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200710020642A Active CN100580691C (en) | 2007-03-16 | 2007-03-16 | Interactive human face identificiating system and method of comprehensive utilizing human face and humanbody auxiliary information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100580691C (en) |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101232571B (en) * | 2008-01-25 | 2010-06-09 | 北京中星微电子有限公司 | Human body image matching method and video analyzing search system |
| CN102201061A (en) * | 2011-06-24 | 2011-09-28 | 常州锐驰电子科技有限公司 | Intelligent safety monitoring system and method based on multilevel filtering face recognition |
| CN102292932A (en) * | 2009-01-23 | 2011-12-21 | 微软公司 | Passive security enforcement |
| CN102332086A (en) * | 2011-06-15 | 2012-01-25 | 夏东 | Facial identification method based on dual threshold local binary pattern |
| CN102446269A (en) * | 2010-10-15 | 2012-05-09 | 微盟电子(昆山)有限公司 | Face recognition method capable of inhibiting noise and environment impact |
| CN102467088A (en) * | 2010-11-16 | 2012-05-23 | 深圳富泰宏精密工业有限公司 | Face recognition alarm clock and method for wakening user by face recognition alarm clock |
| CN102831396A (en) * | 2012-07-23 | 2012-12-19 | 常州蓝城信息科技有限公司 | Computer face recognition method |
| CN102833486A (en) * | 2012-07-05 | 2012-12-19 | 深圳泰山在线科技有限公司 | Method and device for real-time adjusting face display scale in video image |
| CN102981619A (en) * | 2012-11-19 | 2013-03-20 | 广东欧珀移动通信有限公司 | Method and system for replacing theme based on color of user clothes of mobile terminal |
| CN103076879A (en) * | 2012-12-28 | 2013-05-01 | 中兴通讯股份有限公司 | Multimedia interaction method and device based on face information, and terminal |
| CN103119625A (en) * | 2011-09-16 | 2013-05-22 | 华为技术有限公司 | Video character separation method and device |
| WO2013152639A1 (en) * | 2012-04-11 | 2013-10-17 | 腾讯科技（深圳）有限公司 | Video chatting method and system |
| CN103368745A (en) * | 2013-07-19 | 2013-10-23 | 江南大学 | User identity strong authentication method for educational information resource guarantee |
| CN103477350A (en) * | 2011-03-29 | 2013-12-25 | 脸谱公司 | Face recognition based on spatial and temporal proximity |
| CN103646244A (en) * | 2013-12-16 | 2014-03-19 | 北京天诚盛业科技有限公司 | Methods and devices for face characteristic extraction and authentication |
| CN104408404A (en) * | 2014-10-31 | 2015-03-11 | 小米科技有限责任公司 | Face identification method and apparatus |
| CN104517102A (en) * | 2014-12-26 | 2015-04-15 | 华中师范大学 | Method and system for detecting classroom attention of student |
| CN104680053A (en) * | 2013-12-03 | 2015-06-03 | 湖北海洋文化传播有限公司 | Method and device for authenticating identity of current authentication terminal holder |
| CN105631430A (en) * | 2015-12-30 | 2016-06-01 | 浙江宇视科技有限公司 | Matching method and apparatus for face image |
| CN106971133A (en) * | 2016-01-14 | 2017-07-21 | 芋头科技（杭州）有限公司 | One kind improves image recognition precision device and method |
| CN107194363A (en) * | 2017-05-31 | 2017-09-22 | 广东欧珀移动通信有限公司 | image saturation processing method, device, storage medium and computer equipment |
| CN107292252A (en) * | 2017-06-09 | 2017-10-24 | 南京华捷艾米软件科技有限公司 | A kind of personal identification method of autonomous learning |
| CN107563359A (en) * | 2017-09-29 | 2018-01-09 | 重庆市智权之路科技有限公司 | Recognition of face temperature analysis generation method is carried out for dense population |
| CN108154099A (en) * | 2017-12-20 | 2018-06-12 | 北京奇艺世纪科技有限公司 | A kind of character recognition method, device and electronic equipment |
| CN109063596A (en) * | 2018-07-13 | 2018-12-21 | 江苏慧学堂系统工程有限公司 | A kind of face identification system and mobile terminal and recognition of face auxiliary device |
| CN109241886A (en) * | 2018-08-23 | 2019-01-18 | 湖北工业大学 | Face recognition algorithms and system based on OLBP and PCA |
| CN110796079A (en) * | 2019-10-29 | 2020-02-14 | 深圳龙岗智能视听研究院 | Multi-camera visitor identification method and system based on face depth features and human body local depth features |
| CN110929068A (en) * | 2019-11-15 | 2020-03-27 | 南威软件股份有限公司 | Face searching method based on terminal equipment |
| CN111179934A (en) * | 2018-11-12 | 2020-05-19 | 奇酷互联网络科技（深圳）有限公司 | Method of selecting a speech engine, mobile terminal and computer-readable storage medium |
| WO2020113571A1 (en) * | 2018-12-07 | 2020-06-11 | 深圳市欢太科技有限公司 | Face recognition data processing method and apparatus, mobile device and computer readable storage medium |
| CN111339346A (en) * | 2020-02-26 | 2020-06-26 | 深圳市中兴系统集成技术有限公司 | Face recognition system and method based on electronic fence |
| CN111582284A (en) * | 2020-04-27 | 2020-08-25 | 中国科学院信息工程研究所 | Privacy protection method and device for image recognition and electronic equipment |
| CN112036310A (en) * | 2020-08-31 | 2020-12-04 | 北京字节跳动网络技术有限公司 | Picture processing method, device, equipment and storage medium |
| CN112581675A (en) * | 2019-09-27 | 2021-03-30 | 上海慧川机电设备有限公司 | One-card security linkage management system and method |
| CN113111354A (en) * | 2020-01-10 | 2021-07-13 | 北京地平线机器人技术研发有限公司 | Target retrieval method and system, terminal device, cloud server, medium and device |
| CN116434289A (en) * | 2021-12-30 | 2023-07-14 | 同方威视技术股份有限公司 | Millimeter wave or terahertz wave identity verification method and device, millimeter wave or terahertz wave security inspection equipment and electronic equipment |
-
2007
- 2007-03-16 CN CN200710020642A patent/CN100580691C/en active Active
Cited By (56)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101232571B (en) * | 2008-01-25 | 2010-06-09 | 北京中星微电子有限公司 | Human body image matching method and video analyzing search system |
| CN102292932A (en) * | 2009-01-23 | 2011-12-21 | 微软公司 | Passive security enforcement |
| CN102446269B (en) * | 2010-10-15 | 2015-10-14 | 微盟电子(昆山)有限公司 | The recognition algorithms of noise and environmental impact can be suppressed |
| CN102446269A (en) * | 2010-10-15 | 2012-05-09 | 微盟电子(昆山)有限公司 | Face recognition method capable of inhibiting noise and environment impact |
| CN102467088A (en) * | 2010-11-16 | 2012-05-23 | 深圳富泰宏精密工业有限公司 | Face recognition alarm clock and method for wakening user by face recognition alarm clock |
| CN103477350A (en) * | 2011-03-29 | 2013-12-25 | 脸谱公司 | Face recognition based on spatial and temporal proximity |
| US10162999B2 (en) | 2011-03-29 | 2018-12-25 | Facebook, Inc. | Face recognition based on spatial and temporal proximity |
| CN102332086A (en) * | 2011-06-15 | 2012-01-25 | 夏东 | Facial identification method based on dual threshold local binary pattern |
| CN102201061A (en) * | 2011-06-24 | 2011-09-28 | 常州锐驰电子科技有限公司 | Intelligent safety monitoring system and method based on multilevel filtering face recognition |
| CN102201061B (en) * | 2011-06-24 | 2012-10-31 | 常州锐驰电子科技有限公司 | Intelligent safety monitoring system and method based on multilevel filtering face recognition |
| CN103119625B (en) * | 2011-09-16 | 2015-06-03 | 华为技术有限公司 | Video character separation method and device |
| CN103119625A (en) * | 2011-09-16 | 2013-05-22 | 华为技术有限公司 | Video character separation method and device |
| CN103368929B (en) * | 2012-04-11 | 2016-03-16 | 腾讯科技（深圳）有限公司 | A kind of Video chat method and system |
| CN103368929A (en) * | 2012-04-11 | 2013-10-23 | 腾讯科技（深圳）有限公司 | Video chatting method and system |
| WO2013152639A1 (en) * | 2012-04-11 | 2013-10-17 | 腾讯科技（深圳）有限公司 | Video chatting method and system |
| US9094571B2 (en) | 2012-04-11 | 2015-07-28 | Tencent Technology (Shenzhen) Company Limited | Video chatting method and system |
| CN102833486A (en) * | 2012-07-05 | 2012-12-19 | 深圳泰山在线科技有限公司 | Method and device for real-time adjusting face display scale in video image |
| CN102831396A (en) * | 2012-07-23 | 2012-12-19 | 常州蓝城信息科技有限公司 | Computer face recognition method |
| WO2014075430A1 (en) * | 2012-11-19 | 2014-05-22 | 广东欧珀移动通信有限公司 | Method and system for changing theme based on color of clothes of user of mobile terminal |
| CN102981619A (en) * | 2012-11-19 | 2013-03-20 | 广东欧珀移动通信有限公司 | Method and system for replacing theme based on color of user clothes of mobile terminal |
| CN102981619B (en) * | 2012-11-19 | 2015-12-02 | 广东欧珀移动通信有限公司 | A kind of method and system of the color replacement theme based on mobile phone users clothes |
| CN103076879A (en) * | 2012-12-28 | 2013-05-01 | 中兴通讯股份有限公司 | Multimedia interaction method and device based on face information, and terminal |
| CN103368745A (en) * | 2013-07-19 | 2013-10-23 | 江南大学 | User identity strong authentication method for educational information resource guarantee |
| CN104680053A (en) * | 2013-12-03 | 2015-06-03 | 湖北海洋文化传播有限公司 | Method and device for authenticating identity of current authentication terminal holder |
| CN104680053B (en) * | 2013-12-03 | 2018-05-11 | 湖北海洋文化传播有限公司 | To current authentication terminal holder's identity authentication method and device |
| CN103646244A (en) * | 2013-12-16 | 2014-03-19 | 北京天诚盛业科技有限公司 | Methods and devices for face characteristic extraction and authentication |
| CN103646244B (en) * | 2013-12-16 | 2018-01-09 | 北京天诚盛业科技有限公司 | Extraction, authentication method and the device of face characteristic |
| CN104408404A (en) * | 2014-10-31 | 2015-03-11 | 小米科技有限责任公司 | Face identification method and apparatus |
| CN104517102A (en) * | 2014-12-26 | 2015-04-15 | 华中师范大学 | Method and system for detecting classroom attention of student |
| CN104517102B (en) * | 2014-12-26 | 2017-09-29 | 华中师范大学 | Student classroom notice detection method and system |
| CN105631430A (en) * | 2015-12-30 | 2016-06-01 | 浙江宇视科技有限公司 | Matching method and apparatus for face image |
| CN106971133A (en) * | 2016-01-14 | 2017-07-21 | 芋头科技（杭州）有限公司 | One kind improves image recognition precision device and method |
| CN107194363A (en) * | 2017-05-31 | 2017-09-22 | 广东欧珀移动通信有限公司 | image saturation processing method, device, storage medium and computer equipment |
| CN107194363B (en) * | 2017-05-31 | 2020-02-04 | Oppo广东移动通信有限公司 | Image saturation processing method and device, storage medium and computer equipment |
| CN107292252B (en) * | 2017-06-09 | 2020-09-15 | 南京华捷艾米软件科技有限公司 | Identity recognition method for autonomous learning |
| CN107292252A (en) * | 2017-06-09 | 2017-10-24 | 南京华捷艾米软件科技有限公司 | A kind of personal identification method of autonomous learning |
| CN107563359B (en) * | 2017-09-29 | 2018-09-11 | 重庆市智权之路科技有限公司 | Recognition of face temperature is carried out for dense population and analyzes generation method |
| CN107563359A (en) * | 2017-09-29 | 2018-01-09 | 重庆市智权之路科技有限公司 | Recognition of face temperature analysis generation method is carried out for dense population |
| CN108154099A (en) * | 2017-12-20 | 2018-06-12 | 北京奇艺世纪科技有限公司 | A kind of character recognition method, device and electronic equipment |
| CN108154099B (en) * | 2017-12-20 | 2021-04-30 | 北京奇艺世纪科技有限公司 | Figure identification method and device and electronic equipment |
| CN109063596A (en) * | 2018-07-13 | 2018-12-21 | 江苏慧学堂系统工程有限公司 | A kind of face identification system and mobile terminal and recognition of face auxiliary device |
| CN109241886A (en) * | 2018-08-23 | 2019-01-18 | 湖北工业大学 | Face recognition algorithms and system based on OLBP and PCA |
| CN109241886B (en) * | 2018-08-23 | 2021-04-02 | 湖北工业大学 | Face recognition method and system based on OLBP and PCA |
| CN111179934A (en) * | 2018-11-12 | 2020-05-19 | 奇酷互联网络科技（深圳）有限公司 | Method of selecting a speech engine, mobile terminal and computer-readable storage medium |
| WO2020113571A1 (en) * | 2018-12-07 | 2020-06-11 | 深圳市欢太科技有限公司 | Face recognition data processing method and apparatus, mobile device and computer readable storage medium |
| CN112581675A (en) * | 2019-09-27 | 2021-03-30 | 上海慧川机电设备有限公司 | One-card security linkage management system and method |
| CN110796079A (en) * | 2019-10-29 | 2020-02-14 | 深圳龙岗智能视听研究院 | Multi-camera visitor identification method and system based on face depth features and human body local depth features |
| CN110929068B (en) * | 2019-11-15 | 2022-06-21 | 南威软件股份有限公司 | Face searching method based on terminal equipment |
| CN110929068A (en) * | 2019-11-15 | 2020-03-27 | 南威软件股份有限公司 | Face searching method based on terminal equipment |
| CN113111354A (en) * | 2020-01-10 | 2021-07-13 | 北京地平线机器人技术研发有限公司 | Target retrieval method and system, terminal device, cloud server, medium and device |
| CN111339346A (en) * | 2020-02-26 | 2020-06-26 | 深圳市中兴系统集成技术有限公司 | Face recognition system and method based on electronic fence |
| CN111582284A (en) * | 2020-04-27 | 2020-08-25 | 中国科学院信息工程研究所 | Privacy protection method and device for image recognition and electronic equipment |
| CN111582284B (en) * | 2020-04-27 | 2023-04-07 | 中国科学院信息工程研究所 | Privacy protection method and device for image recognition and electronic equipment |
| CN112036310A (en) * | 2020-08-31 | 2020-12-04 | 北京字节跳动网络技术有限公司 | Picture processing method, device, equipment and storage medium |
| US11900726B2 (en) | 2020-08-31 | 2024-02-13 | Beijing Bytedance Network Technology Co., Ltd. | Picture processing method and apparatus, device, and storage medium |
| CN116434289A (en) * | 2021-12-30 | 2023-07-14 | 同方威视技术股份有限公司 | Millimeter wave or terahertz wave identity verification method and device, millimeter wave or terahertz wave security inspection equipment and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100580691C (en) | 2010-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100580691C (en) | Interactive human face identificiating system and method of comprehensive utilizing human face and humanbody auxiliary information | |
| CN109993160B (en) | Image correction and text and position identification method and system | |
| Chen et al. | Relevance metric learning for person re-identification by exploiting listwise similarities | |
| Xie et al. | m-SNE: Multiview stochastic neighbor embedding | |
| Bedagkar-Gala et al. | Multiple person re-identification using part based spatio-temporal color appearance model | |
| CN102682309B (en) | Face feature registering method and device based on template learning | |
| CN103577818A (en) | Method and device for recognizing image characters | |
| CN112464730B (en) | Pedestrian re-identification method based on domain-independent foreground feature learning | |
| CN113515988A (en) | Palm print recognition method, feature extraction model training method, device and medium | |
| CN107273783A (en) | Face identification system and its method | |
| Voronov et al. | Faces 2D-recognition аnd identification using the HOG descriptors method | |
| Bruns et al. | Mobile phone-enabled museum guidance with adaptive classification | |
| CN109740417A (en) | Invoice type recognition methods, device, storage medium and computer equipment | |
| Amaro et al. | Evaluation of machine learning techniques for face detection and recognition | |
| CN108960186B (en) | Advertising machine user identification method based on human face | |
| Tahmasebi et al. | Robust intra-class distance-based approach for multimodal biometric game theory-based rank-level fusion of ear, palmprint and signature | |
| CN117333908A (en) | Cross-modal pedestrian re-recognition method based on attitude feature alignment | |
| CN108830217B (en) | Automatic signature distinguishing method based on fuzzy mean hash learning | |
| CN116434344A (en) | Pen holding gesture recognition model training method, recognition method and related devices | |
| CN110909678A (en) | Face recognition method and system based on width learning network feature extraction | |
| CN115601791A (en) | Unsupervised pedestrian re-identification method based on Multiformer and outlier sample re-distribution | |
| Liu et al. | An iris recognition approach with SIFT descriptors | |
| CN113901959A (en) | Authentication method based on RFID single-label non-contact gesture recognition system | |
| Kawanishi et al. | Which content in a booklet is he/she reading? Reading content estimation using an indoor surveillance camera | |
| Impedovo et al. | Stability analysis of static signatures for automatic signature verification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20090327 Address after: Jiangsu province Nanjing city Guangzhou Road No. 5 building 2308 room 1 Dragon International Postal encoding: 210008 Applicant after: Nanjing bright Video Technology Co., Ltd. Address before: Floor 20, civil defense building, 189 Guangzhou Road, Gulou District, Jiangsu, Nanjing Province, China: 210029 Applicant before: Nanjing Seekpai Information Technology Co., Ltd. |
|
| ASS | Succession or assignment of patent right |
Owner name: NANJING MINGMAI SHIXUN SCIENCE CO., LTD. Free format text: FORMER OWNER: NANJING SOUPAI INFORMATION TECHNOLOGY CO., LTD. Effective date: 20090327 |
|
| ASS | Succession or assignment of patent right |
Owner name: SHANGHAI BOKANG INTELLIGENT INFORMATION TECHNOLOGY Free format text: FORMER OWNER: NANJING MINGMAI SHIXUN SCIENCE CO., LTD. Effective date: 20090925 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20090925 Address after: Shanghai City, Guangxi No. 69 building 3 floor Jinglu 27 post encoding: 200233 Applicant after: Shanghai Bokang Intelligent Information Technology Co., Ltd. Address before: Jiangsu province Nanjing city Guangzhou Road No. 5 building 2308 room 1 Dragon International Postal encoding: 210008 Applicant before: Nanjing bright Video Technology Co., Ltd. |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |