CN105389548A - Love and marriage evaluation system and method based on face recognition - Google Patents
Love and marriage evaluation system and method based on face recognition Download PDFInfo
- Publication number
- CN105389548A CN105389548A CN201510697448.8A CN201510697448A CN105389548A CN 105389548 A CN105389548 A CN 105389548A CN 201510697448 A CN201510697448 A CN 201510697448A CN 105389548 A CN105389548 A CN 105389548A
- Authority
- CN
- China
- Prior art keywords
- marriage
- face
- similarity
- love
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a love and marriage evaluation system and a love and marriage evaluation method based on face recognition. The love and marriage evaluation method comprises the steps of: reading a wedding picture of a man and a woman from a wedding picture database, and inputting the wedding picture to a face recognition system; filtering noise in the image; performing contrast enhancement on the whole image, and enhancing signals of a main body part; performing normalization processing on the image, and reducing image signal variation of main body feature part of the image caused by factors such as illumination; positioning faces of the man and the woman in the whole picture, and extracting fixed positions of facial features; extracting a facial feature set; matching the obtained facial feature set of the picture with objects in the picture database to achieve tagging of a feature set; reading one feature set from a training sample, and calculating the similarity between a tested sample and the training sample; traversing the whole training sample set to obtain all similarity values, ranking the similarity values in a reversed manner, and selecting the training sample with the highest similarity for subscript and output; calculating a love and marriage index value of the tested sample; and outputting the love and marriage index obtained through calculation in an interval form.
Description
Technical field
The present invention relates to a kind of marriage emotion assessment method and system, particularly a kind of love and marriage evaluation system based on recognition of face and method.
Background technology
Within 2014, Chinese marriage and making friend's overall market size is 81.8 hundred million yuans, sequential growth rate 2.0%, estimates that love and marriage in end of the year market in 2015 will reach 83,100,000,000 yuans.The further raising of love and marriage acceptance of the O2O form that combines with solid shop/brick and mortar store to network with the unmarried crowd of being old enough to get married of China, Future Internet marriage and making friend also can progressively improve in the permeability in overall love and marriage market.From love and marriage market, internet in 2014 income pattern analysis, century good edge, lily net, treasure net, have hoddy four to occupy the total share in market of 70% altogether
Century, good edge started love and marriage proposed algorithm project as the internet love and marriage platform that the whole nation is maximum from 2011, and initial trial uses the method for Item-Based (based on commodity) to construct a Man-Woman (man-woman) matrix.If their a hypothesis man likes a woman, so he must like the woman similar with this woman.By woman as commodity, using the birthday of woman, height, body weight, birthplace, educational background, work, hobby, character type, income, expense etc. as base attribute, calculate the similarity between woman, recommend man.Be easy to like this cause a result, each man likes the boudoir honey of oneself woman.
Above-mentioned existing method exists following not enough, specifically comprises: 1) unidirectional recommendation women is to male sex's problem: that many times show is beauty, cause a small amount of women to receive the letter of most of male sex, and most of women can not receive.Also cannot recommend the male sex to women simultaneously.2) Sparse Problem: women welcome guest's flag data is abundant not, or female group skewness, all can affect accuracy rate and the recall rate of recommendation results.3) traditional problem of collaborative filtering: parameter dimensions is more, calculates more complicated.Later stage have employed the ReciprocalRecommendation algorithm of bidirectional relationship for this reason, obtain Man->Women respectively, Woman->Man, is then multiplied or matching solves the problems referred to above 1, but still cannot solve the problem 2 and 3.
The saying that ancient Chinese has " man and wife's phase ", refer to that face is mutually similar, soul communicates, and is accustomed to convergent, interacts.Biologist thinks that the people feeling similar to oneself appearance in subconsciousness is more credible.Man and wife is divided into similar type, alike type, Qin Renxing, comprehensive, complementary type, Mars collision type, syntonic type, response type, relationship type by some articles mutually.But, there is no the quantification sentiment analysis method and system of any scholar proposition based on face phase at present.
In sum, the existing collaborative recommendation method based on user property label data, is needed just can be recommended by the test of a large amount of question and answer, not only wastes time and energy, also there is the problem such as Sparse and computation complexity.And the first impression analysis in human communication does not carry out image domains quantitative analysis yet.And the present invention can solve problem above well.
Summary of the invention
It is main men and women's love and marriage collaborative recommendation method and system based on personal information text label that the object of the invention there are provided a kind of, solve based on the restriction of single structure fractional analysis, analyze the problem of the face feature love and marriage index of men and women from non-structured view data, realize the functions such as love and marriage affection index judgement fast
The present invention solves the technical scheme that its technical matters takes: a kind of love and marriage evaluation system based on recognition of face, and this system comprises marriage registration image data base, marriage registration number tag database, image denoising strengthen module, normalization module, face's edge detection module, face feature collection acquisition module, men and women's feature set relationships quantify module, similarity calculation module, love and marriage index computing module.
The function of marriage registration image data base is: for image recognition and analysis.
The function of marriage registration number tag database is: comprise the marriage date or incidentally divorce the date, for demarcating personage's affection index.
Image denoising strengthens the function of module: removal of images with noise, reduce image information isolated point noise spot to the impact of post-processed, strengthen the accuracy of image detection algorithm.
The function of normalization module is: be normalized the size of face and color, and size, angle as brought face size, distance of taking pictures correct; To the image under different shooting condition and the image through exposure-processed, carry out gray scale normalization and obtain stable standard picture;
The function of face's edge detection module is: the face mask obtaining men and women from image, and intercepts out from whole image by face image.
The function of face feature collection acquisition module is: the feature set gathering each principal character organ of face from face image inside, as information such as eyebrow, eyes, nose, forehead, cheek, mouth, ear, hair styles, for follow-up machine learning model provides set of eigenvectors;
The function of men and women's feature set relationships quantify module is: with reference to the information in existing marriage database, calculates men and women and to marry the time limit, quantize men and women's affection index;
The function of similarity calculation module is: calculate the feature set of the group photo of required test men and women and the characteristic similarity in marriage registration image data base;
The function of love and marriage index computing module is: the sample the most similar to marriage registration image data base according to test men and women group photo, calculates the love and marriage index of test sample book in conjunction with the sample marriage time limit.
Present invention also offers a kind of implementation method of the love and marriage evaluation system based on recognition of face, the method comprises the steps:
Step 1: read men and women and marry photo from marriage photo database, be input to face identification system.
Step 2: adopt image de-noising method to worry to the noise in image.
Step 3: carry out Contrast enhanced to whole image, strengthens main part signal.
Step 4: be normalized image, weakens the image subject characteristic picture signal change that the factors such as illumination cause.
Step 5: the face of locating men and women from whole picture, for face feature extracts fixed position.
Step 6: extract people face characteristic set, comprise the length of eyebrow and span profile, glabella apart from, angle of inclination, eye contour, profile, two spacing, nose profile, naris position and size, forehead length and width and the accounting in whole facial contour, the characteristic information such as length and width, lip thickness, corners of the mouth degree of tilt of mouth.
Step 7: by the object matching in the photo face feature collection of acquisition in step 6 and picture data storehouse, realization character collection labeling.
Step 8: read a feature set from training sample, calculates the similarity between test sample book and training sample.
Step 9: travel through whole training sample set (that is: marriage certificate picture data storehouse), obtain all Similarity value, and carry out down sequence, select the highest training sample subscript of similarity and export.
Step 10: according to the marriage time limit in similar training sample 1 marriage registration database, calculates test sample book love and marriage exponential quantity.Step 11: form interval for the love and marriage index calculated is exported.
Effective effect:
1, the present invention adopts face recognition technology, by carrying out face characteristic collection to marriage registration image data base, builds the men and women's love and marriage Exponential Sample space based on face feature collection in conjunction with marriage registration structured database.
2, test sample book of the present invention is by calculating the face characteristic similarity with men and women in database, calculate the love and marriage index range of sample, be supplementing the structure based Collaborative Filtering algorithm of existing online love and marriage platform, set forth the idea of " man and wife's phase " of traditional metaphysics from image recognition technology angle.
3, the present invention introduces dual similar function, breaks through existing single recognition of face and retrieval technique, proposes the overall similarity computation model of men and women, introduces hyperbolic function computing method simultaneously, solves the difficult problem that general Similarity Measure cannot provide confidence space.
Accompanying drawing explanation
Fig. 1 is that the wedding photo original drawings that the present invention uses illustrate intention.
Fig. 2 is that wedding photo organ contours feature set puies forward result schematic diagram.
Fig. 3 is system architecture diagram of the present invention.
Fig. 4 is method flow diagram of the present invention.
Embodiment
Below in conjunction with Figure of description, the invention is described in further detail.
As shown in Figure 3, the present invention discloses a kind of love and marriage evaluation system based on recognition of face, and this system comprises marriage registration image data base, marriage registration number tag database, image denoising enhancing module, normalization module, face's edge detection module, face feature collection acquisition module, men and women's feature set relationships quantify module, similarity calculation module, love and marriage index computing module.
The function of marriage registration image data base is: for image recognition and analysis.
The function of marriage registration number tag database is: comprise the marriage date or incidentally divorce the date, for demarcating personage's affection index.
Image denoising strengthens the function of module: removal of images with noise, reduce image information isolated point noise spot to the impact of post-processed, strengthen the accuracy of image detection algorithm.
The function of normalization module is: be normalized the size of face and color, and size, angle as brought face size, distance of taking pictures correct; To the image under different shooting condition and the image through exposure-processed, carry out gray scale normalization and obtain stable standard picture;
The function of face's edge detection module is: the face mask obtaining men and women from image, and intercepts out from whole image by face image.
The function of face feature collection acquisition module is: the feature set gathering each principal character organ of face from face image inside, as information such as eyebrow, eyes, nose, forehead, cheek, mouth, ear, hair styles, for follow-up machine learning model provides set of eigenvectors;
The function of men and women's feature set relationships quantify module is: with reference to the information in existing marriage database, calculates men and women and to marry the time limit, quantize men and women's affection index;
The function of similarity calculation module is: calculate the feature set of the group photo of required test men and women and the characteristic similarity in marriage registration image data base;
The function of love and marriage index computing module is: the sample the most similar to marriage registration image data base according to test men and women group photo, calculates the love and marriage index of test sample book in conjunction with the sample marriage time limit.
The marriage photo of registry shooting, is generally red background as shown in Figure 1, and the head photo of men and women shines based on face front, not wearing flash of light pendant part, without lens glasses (that is: only having picture frame).
Single organ contours feature extraction intermediate result figure as shown in Figure 2, if the length of eyebrow and span profile, glabella are apart from, angle of inclination, eye contour, profile, two spacing, nose profile, naris position and size, forehead length and width and the accounting in whole facial contour, characteristic information the presenting on master pattern such as length and width, lip thickness, corners of the mouth degree of tilt of mouth.
System is made up of data Layer and identification layer as shown in Figure 3.Data Layer comprises marriage registration database, marriage license sheet database and test sample book.Identification layer comprises denoising module, normalization module, edge detection module, collection apparatus module, similarity calculation module, men and women's characteristic quantification module and index computing module.
Implementing procedure of the present invention as shown in Figure 4, comprise a marriage registration database, marriage picture data and test sample book photograph collection three source datas, comprise image color change module, gray scale normalization module, image denoising enhancing module, face's edge detection module, face feature acquisition module, similarity calculation module and love and marriage index computing module.Idiographic flow is as follows:
Step 1: read men and women one by one and marry photo from marriage photo database, as the original input of system, be input to face identification system.
Step 2: due to the difference of photograph taking process and preserving type, a lot of photo exists noise.This step adopts conventional denoising method, as methods such as linear filtering, medium filtering, Wiener filterings, filters the noise existed in image.
Step 3: navigate to face to be more prone in the picture, adds image comparison and strengthens algorithm, strengthens main part as adopted the method such as histogram equalization or serpentine change.
Step 4: the normalized of carrying out facial image, comprises geometrical normalization and gray scale normalization.Geometrical normalization comprises the normalization of face size, plane face rotational correction, degree of depth face rotational correction three links.Gray scale normalization is used for compensating the facial image obtained under different light intensity, light source direction, changes to weaken the picture signal caused due to illumination variation merely.
Step 5: the face of locating men and women from whole picture.Consider that the backcolor of marriage registration photograph is single, profile and the colour of skin can be used as important information.Have more the similarity of skin pixel on color table and correlativity is spatially partitioned into possible human face region.
Step 6: extract people face characteristic set.Consider that marriage registration is full face according to majority, better, comparatively greatly, the present invention adopts the face identification method based on geometric properties to facial image effect.Detect face feature point, by detecting the relative distance between these key points, obtain the eigenvector (as Fig. 2) describing each face, that is: the length of eyebrow and span profile, glabella are apart from, angle of inclination, eye contour, profile, two spacing, nose profile, naris position and size, forehead length and width and the accounting in whole facial contour, the characteristic information such as length and width, lip thickness, corners of the mouth degree of tilt of mouth.And the relation calculated between these features.Obtain training sample characteristic set space
wherein
with
represent masculinity and femininity i-th facial eigenvectors value in training sample respectively
Step 7: by test sample book photo input face identification system, obtain the to be tested object characteristic the same with marriage certificate picture data storehouse
wherein
with
represent masculinity and femininity i-th facial eigenvectors value in test sample book respectively.
Step 8: read a feature set from training sample, calculates the similarity between test sample book and training sample.Consider that comparison target of the present invention is men and women's two targets, and there is the difference of essence in traditional individual human face recognition technology.Propose the multiobject image similarity algorithm of core at this, provide two kinds of computing functions at this.
Method one: adopt traditional face feature matching algorithm, calculate the similarity of men and women respectively, obtain face feature similarity Sim
m(male sex's similarity) and Sim
f(women's similarity), then gets the product square root of two similarities, that is:
Method two: adopt feature to match pattern, calculate comprehensive similarity, the feature space as training sample is
the feature space of test sample book is
then similarity function is:
Step 9: travel through whole training sample set (marriage certificate picture data storehouse), obtain all Similarity value, and do down sequence, selects the highest training sample subscript of similarity and exports.Consider and may occur front several similarity all relatively, the present invention provides a kind of rank and intercepts algorithm.If similarity rank is { Sim
1, Sim
2..., Sim
n), if
(wherein Δ is gradient decent degree, can be set to 10% or 5%), then think that the training sample 1 after sequence has absolute similarity degree with test sample book, and in this, as unique Similarity matching.
then think that the training sample 1 and 2 after sequence can be used for describing test sample book.Gradient decent degree after calculating sequence by that analogy between sample 2 and sample 3 intercepts.If p time iteration successor is so less than Δ, then get the training sample after p sequence as output.
Step 10: with reference to the marriage time limit in similar training sample 1 marriage registration database, calculates test sample book love and marriage exponential quantity.If training sample marriage registration is My, divorce registration is Dy, then the love and marriage index M vs=Dy-My of training sample.If there is no divorce records, then Mvs=100.If the training sample of p sequence, then calculate the Mvt of this p training sample, and ask the weighted mean value of all Mvt as new Mvt.If similarity is 1, then the Mvt=Mvs of test sample book.Consider that interval that Similarity Measure obtains is between [01], the present invention proposes hyperbolic function and solves acquisition love and marriage index codomain [Mvt-Δ Mvt+ Δ], wherein
Step 11: form interval for the love and marriage index calculated is exported, as [3; 5; 0.8] to represent and crowd that tested object has in the marriage object of similar face phase 80% maintains the marital relations of 3 ~ 5 years.
In sum, the present invention is directed in existing love and marriage commending system the information simplification problem using single text label information to adopt Collaborative Filtering Recommendation Algorithm, propose a kind of love and marriage index number evaluation method based on face recognition technology.The multiple goal similarity calculation that the present invention proposes, not only may be used for men and women's love and marriage Similarity Measure, can also be used for the application scenarioss such as his many personages co-occurrence discovery.
The foregoing is only preferred case study on implementation of the present invention, be not limited to the present invention, although with reference to previous embodiment to invention has been detailed description, for a person skilled in the art, it still can improve the technical scheme described in foregoing embodiments, or replaces on an equal basis wherein portion of techniques.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (6)
1. the love and marriage evaluation system based on recognition of face, its spy is being, described system comprises: marriage registration image data base, marriage registration number tag database, image denoising strengthen module, normalization module, face's edge detection module, face feature collection acquisition module, men and women's feature set relationships quantify module, similarity calculation module, love and marriage index computing module;
The function of marriage registration image data base is: for image recognition and analysis;
The function of marriage registration number tag database is: comprise the marriage date or incidentally divorce the date, for demarcating personage's affection index;
Image denoising strengthens the function of module: removal of images with noise, reduce image information isolated point noise spot to the impact of post-processed, strengthen the accuracy of image detection algorithm;
The function of normalization module is: be normalized the size of face and color, and size, angle as brought face size, distance of taking pictures correct; To the image under different shooting condition and the image through exposure-processed, carry out gray scale normalization and obtain stable standard picture;
The function of face's edge detection module is: the face mask obtaining men and women from image, and intercepts out from whole image by face image;
The function of face feature collection acquisition module is: the feature set gathering each principal character organ of face from face image inside, the i.e. information such as eyebrow, eyes, nose, forehead, cheek, mouth, ear, hair style, for follow-up machine learning model provides set of eigenvectors;
The function of men and women's feature set relationships quantify module is: with reference to the information in existing marriage database, calculates men and women and to marry the time limit, quantize men and women's affection index;
The function of similarity calculation module is: calculate the feature set of the group photo of required test men and women and the characteristic similarity in marriage registration image data base;
The function of love and marriage index computing module is: the sample the most similar to marriage registration image data base according to test men and women group photo, calculates the love and marriage index of test sample book in conjunction with the sample marriage time limit.
2., based on an implementation method for the love and marriage evaluation system of recognition of face, the method comprises the steps:
Step 1: read men and women one by one and marry photo from marriage photo database, as the original input of system, be input to face identification system;
Step 2: adopt conventional denoising method, as methods such as linear filtering, medium filtering, Wiener filterings, the noise existed in image is filtered;
Step 3: add image comparison and strengthen algorithm, adopts the method for histogram equalization or serpentine change to strengthen main part;
Step 4: the normalized of carrying out facial image, comprises geometrical normalization and gray scale normalization; Geometrical normalization comprises the normalization of face size, plane face rotational correction, degree of depth face rotational correction three links; Gray scale normalization is used for compensating the facial image obtained under different light intensity, light source direction, changes to weaken the picture signal caused due to illumination variation merely;
Step 5: the face of locating men and women from whole picture, uses profile and the colour of skin as important information, has more the similarity of skin pixel on color table and correlativity is spatially partitioned into possible human face region;
Step 6: extract people face characteristic set; Adopt the face identification method based on geometric properties, detect face feature point, by detecting the relative distance between these key points, obtain the eigenvector describing each face, that is: the length of eyebrow and span profile, glabella are apart from, angle of inclination, eye contour, profile, two spacing, nose profile, naris position and size, forehead length and width and the accounting in whole facial contour, the characteristic information such as length and width, lip thickness, corners of the mouth degree of tilt of mouth, and the relation calculated between these features, obtain training sample characteristic set space
wherein
with
represent masculinity and femininity i-th facial eigenvectors value in training sample respectively;
Step 7: by test sample book photo input face identification system, obtain the to be tested object characteristic the same with marriage certificate picture data storehouse
wherein
with
represent masculinity and femininity i-th facial eigenvectors value in test sample book respectively;
Step 8: read a feature set from training sample, calculates the similarity between test sample book and training sample;
Step 9: travel through whole training sample set, i.e. marriage certificate picture data storehouse, obtain all Similarity value, and do down sequence, selects the highest training sample subscript of similarity and export;
Step 10: according to the marriage time limit in the marriage registration database of similar training, calculates test sample book love and marriage exponential quantity; Step 11: form interval for the love and marriage index calculated is exported.
3. the implementation method of a kind of love and marriage evaluation system based on recognition of face according to claim 2, it is characterized in that, the step 8 of described method calculates the similarity between test sample book and training sample, is the method for two kinds of computing functions, comprises:
Method one: adopt traditional face feature matching algorithm, calculate the similarity of men and women respectively, obtain face feature similarity Sim
m(male sex's similarity) and Sim
f(women's similarity), then gets the product square root of two similarities, that is:
Method two: adopt feature to match pattern, calculate comprehensive similarity, the feature space as training sample is
the feature space of test sample book is
then similarity function is:
4. the implementation method of a kind of love and marriage evaluation system based on recognition of face according to claim 2, it is characterized in that, the step 9 of described method comprises: consider and may occur front several similarity all relatively, and described method provides a kind of rank and intercepts algorithm, if similarity rank is { Sim
1, Sim
2..., Sim
nif
wherein Δ is gradient decent degree, is set to 10% or 5%, then think that the training sample 1 after sequence has absolute similarity degree with test sample book, and in this, as unique Similarity matching;
then think that the training sample 1 and 2 after sequence can be used for describing test sample book; Gradient decent degree after calculating sequence by that analogy between sample 2 and sample 3 intercepts, if p iteration successor is so less than Δ, then gets the training sample after p sequence as output.
5. the implementation method of a kind of love and marriage evaluation system based on recognition of face according to claim 2, it is characterized in that, the step 10 of described method comprises: training sample marriage registration is My, and divorce registration is Dy, then the love and marriage index M vs=Dy-My of training sample; If do not have divorce records, then Mvs=100, if the training sample of p sequence, then calculates the Mvt of this p training sample, and asks the weighted mean value of all Mvt as new Mvt, if similarity is 1, then and the Mvt=Mvs of test sample book; Consider that interval that Similarity Measure obtains is between [01], the present invention proposes hyperbolic function and solves acquisition love and marriage index codomain [Mvt-Δ Mvt+ Δ], wherein
6. the implementation method of a kind of love and marriage evaluation system based on recognition of face according to claim 2, it is characterized in that, the step 11 of described method comprises: described love and marriage index [3; 5; 0.8] to represent and crowd that tested object has in the marriage object of similar face phase 80% maintains the marital relations of 3 ~ 5 years.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510697448.8A CN105389548A (en) | 2015-10-23 | 2015-10-23 | Love and marriage evaluation system and method based on face recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510697448.8A CN105389548A (en) | 2015-10-23 | 2015-10-23 | Love and marriage evaluation system and method based on face recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105389548A true CN105389548A (en) | 2016-03-09 |
Family
ID=55421820
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510697448.8A Pending CN105389548A (en) | 2015-10-23 | 2015-10-23 | Love and marriage evaluation system and method based on face recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105389548A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105931327A (en) * | 2016-04-15 | 2016-09-07 | 广东京奥信息科技有限公司 | Access control monitoring method and system |
CN106021455A (en) * | 2016-05-17 | 2016-10-12 | 乐视控股(北京)有限公司 | Image characteristic relationship matching method, apparatus and system |
CN106778500A (en) * | 2016-11-11 | 2017-05-31 | 北京小米移动软件有限公司 | A kind of method and apparatus for obtaining people's object plane phase information |
CN108038496A (en) * | 2017-12-04 | 2018-05-15 | 华南师范大学 | Love and marriage object matching data processing method, device, computer equipment and storage medium based on big data and deep learning |
CN108154165A (en) * | 2017-11-20 | 2018-06-12 | 华南师范大学 | Love and marriage object matching data processing method, device, computer equipment and storage medium based on big data and deep learning |
CN108463823A (en) * | 2016-11-24 | 2018-08-28 | 华为技术有限公司 | A kind of method for reconstructing, device and the terminal of user's Hair model |
CN109685892A (en) * | 2018-12-31 | 2019-04-26 | 南京邮电大学盐城大数据研究院有限公司 | A kind of quick 3D face building system and construction method |
CN110222026A (en) * | 2016-05-24 | 2019-09-10 | 甘肃百合物联科技信息有限公司 | A kind of constructive memory network and the method for being used for prediction mood |
CN110837600A (en) * | 2019-10-12 | 2020-02-25 | 惠州市德赛西威汽车电子股份有限公司 | Intelligent marriage and love matching method based on driving data |
CN113792672A (en) * | 2021-09-16 | 2021-12-14 | 平安科技(深圳)有限公司 | Public place health code acquisition method, device, equipment and medium |
CN114287017A (en) * | 2019-08-29 | 2022-04-05 | 富士通株式会社 | Pattern extraction program, device and method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101847144A (en) * | 2009-03-27 | 2010-09-29 | 上海薇艾信息科技有限公司 | Portrait processing method for Internet dating |
US20120114197A1 (en) * | 2010-11-09 | 2012-05-10 | Microsoft Corporation | Building a person profile database |
CN103678394A (en) * | 2012-09-21 | 2014-03-26 | 孟露芳 | Image matching degree based marriage dating recommendation method and system |
-
2015
- 2015-10-23 CN CN201510697448.8A patent/CN105389548A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101847144A (en) * | 2009-03-27 | 2010-09-29 | 上海薇艾信息科技有限公司 | Portrait processing method for Internet dating |
US20120114197A1 (en) * | 2010-11-09 | 2012-05-10 | Microsoft Corporation | Building a person profile database |
CN103678394A (en) * | 2012-09-21 | 2014-03-26 | 孟露芳 | Image matching degree based marriage dating recommendation method and system |
Non-Patent Citations (1)
Title |
---|
贺笑: ""基于形状特性的人脸相似度的描述与分析"", 《中国优秀硕士学位论文全文数据库 (电子期刊) 信息科技辑》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105931327A (en) * | 2016-04-15 | 2016-09-07 | 广东京奥信息科技有限公司 | Access control monitoring method and system |
CN106021455A (en) * | 2016-05-17 | 2016-10-12 | 乐视控股(北京)有限公司 | Image characteristic relationship matching method, apparatus and system |
WO2017197826A1 (en) * | 2016-05-17 | 2017-11-23 | 乐视控股(北京)有限公司 | Relationship-based image feature matching method, device, and system |
CN110222026A (en) * | 2016-05-24 | 2019-09-10 | 甘肃百合物联科技信息有限公司 | A kind of constructive memory network and the method for being used for prediction mood |
CN110222026B (en) * | 2016-05-24 | 2021-03-02 | 甘肃百合物联科技信息有限公司 | Method for constructing memory network and predicting emotion by using memory network |
CN106778500A (en) * | 2016-11-11 | 2017-05-31 | 北京小米移动软件有限公司 | A kind of method and apparatus for obtaining people's object plane phase information |
CN106778500B (en) * | 2016-11-11 | 2019-09-17 | 北京小米移动软件有限公司 | A kind of method and apparatus obtaining personage face phase information |
CN108463823A (en) * | 2016-11-24 | 2018-08-28 | 华为技术有限公司 | A kind of method for reconstructing, device and the terminal of user's Hair model |
CN108154165B (en) * | 2017-11-20 | 2021-12-07 | 华南师范大学 | Marriage and love object matching data processing method and device based on big data and deep learning, computer equipment and storage medium |
CN108154165A (en) * | 2017-11-20 | 2018-06-12 | 华南师范大学 | Love and marriage object matching data processing method, device, computer equipment and storage medium based on big data and deep learning |
CN108038496A (en) * | 2017-12-04 | 2018-05-15 | 华南师范大学 | Love and marriage object matching data processing method, device, computer equipment and storage medium based on big data and deep learning |
CN109685892A (en) * | 2018-12-31 | 2019-04-26 | 南京邮电大学盐城大数据研究院有限公司 | A kind of quick 3D face building system and construction method |
CN114287017A (en) * | 2019-08-29 | 2022-04-05 | 富士通株式会社 | Pattern extraction program, device and method |
CN110837600A (en) * | 2019-10-12 | 2020-02-25 | 惠州市德赛西威汽车电子股份有限公司 | Intelligent marriage and love matching method based on driving data |
CN113792672A (en) * | 2021-09-16 | 2021-12-14 | 平安科技(深圳)有限公司 | Public place health code acquisition method, device, equipment and medium |
CN113792672B (en) * | 2021-09-16 | 2024-05-28 | 平安科技(深圳)有限公司 | Public place health code acquisition method, device, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105389548A (en) | Love and marriage evaluation system and method based on face recognition | |
CN107273876B (en) | A kind of micro- expression automatic identifying method of ' the macro micro- transformation model of to ' based on deep learning | |
CN108615010B (en) | Facial expression recognition method based on parallel convolution neural network feature map fusion | |
CN108985377B (en) | A kind of image high-level semantics recognition methods of the multiple features fusion based on deep layer network | |
CN104850825B (en) | A kind of facial image face value calculating method based on convolutional neural networks | |
CN106778584B (en) | A kind of face age estimation method based on further feature Yu shallow-layer Fusion Features | |
Laurentini et al. | Computer analysis of face beauty: A survey | |
CN112132833B (en) | Dermatological image focus segmentation method based on deep convolutional neural network | |
CN101819628B (en) | Method for performing face recognition by combining rarefaction of shape characteristic | |
Scandrett et al. | A person-specific, rigorous aging model of the human face | |
CN101833672B (en) | Sparse representation face identification method based on constrained sampling and shape feature | |
CN101833654B (en) | Sparse representation face identification method based on constrained sampling | |
CN110197729A (en) | Tranquillization state fMRI data classification method and device based on deep learning | |
CN104794693A (en) | Human image optimization method capable of automatically detecting mask in human face key areas | |
CN103268485A (en) | Sparse-regularization-based face recognition method capable of realizing multiband face image information fusion | |
CN112465905A (en) | Characteristic brain region positioning method of magnetic resonance imaging data based on deep learning | |
CN107392151A (en) | Face image various dimensions emotion judgement system and method based on neutral net | |
CN111062260B (en) | Automatic generation method of face-beautifying recommendation scheme | |
CN109255289A (en) | A kind of across aging face identification method generating model based on unified formula | |
Ilmini et al. | Computational personality traits assessment: A review | |
CN111950430A (en) | Color texture based multi-scale makeup style difference measurement and migration method and system | |
CN102495999A (en) | Face recognition method | |
CN111666845A (en) | Small sample deep learning multi-mode sign language recognition method based on key frame sampling | |
Guo et al. | Multifeature extracting CNN with concatenation for image denoising | |
CN108985301A (en) | A kind of hyperspectral image classification method of the sub- dictionary learning of multiple features class |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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: 20160309 |