CN103294982A - Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network - Google Patents
Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network Download PDFInfo
- Publication number
- CN103294982A CN103294982A CN2012100427898A CN201210042789A CN103294982A CN 103294982 A CN103294982 A CN 103294982A CN 2012100427898 A CN2012100427898 A CN 2012100427898A CN 201210042789 A CN201210042789 A CN 201210042789A CN 103294982 A CN103294982 A CN 103294982A
- Authority
- CN
- China
- Prior art keywords
- face
- sorter
- people
- age
- pedestrian
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method and system for figure detection, body part positioning, age estimation and gender identification in a picture of a network. The method includes the steps that by means of a face detection technique based on Haar-Like characteristics and Adaboost trainings and a pedestrian detection technique based on HOG characteristics and SVM trainings, a face area and an overall body area are primarily detected, and then a further fusion is carried out so that a face can correspond to a body if the body exists; active appearance models (AAM) are used in the face part so as to position the positions of the five sense organs of the face, and the position of the upper half body, the position of the lower half body, the positions of the left hand and the right hand, and the position of feet are confirmed on a body part according to a human body geometric model; eventually, identification of gender and age is carried out on the face part according to GIST characteristics and an SVM. Therefore, whether a figure exists in the picture of the network and various biological characteristics of the figure can be obtained through calculation.
Description
Technical field
The present invention relates to the method and system of person detecting in a kind of network picture, body part location, age estimation and sex identification, belong to field of image recognition, be in static images, to carry out the detection of people's face and the detection of upright whole body specifically, locate the position of hair, eyes, nose, cheek and face then in people face position, and estimate age and the sex of people's face; Position the body part location upper part of the body, the lower part of the body, foot and right-hand man.
Background technology
Picture in the internet is based on the personage, no matter be news picture or commodity picture, and relevant with the people overwhelming majority of occupying, that is to say in the most picture has the personage to occur.Wherein people's face or pedestrian detection can be come out by human face detection tech and pedestrian detection technology, still, for the not more research in the refinement location at each position.The present invention is based on human face detection tech and pedestrian detection technology, human face region and health overall region are detected at first tentatively, do further fusion again, make the corresponding health of people's face, if there is health; Use AAM(Active Appearance Models, active contour model then in people face position) location face face position, determine the upper part of the body, the lower part of the body, right-hand man and foot position at body part according to human geometry's model; At last carry out the identification of age and sex in people face position; So, whether one throws the net exists the personage in the network picture, and the personage's who exists various biological characteristics just calculate.
Summary of the invention
The method and system that the purpose of this invention is to provide person detecting in a kind of network picture, body part location, age estimation and sex identification, comprise: to people's face sample training people face sorter, to pedestrian's sample training pedestrian sorter, to people's face sample training AAM model data, to people's face sample training age in different sexes all ages and classes stage and the sorter of sex identification, load sorter separately then, be implemented in person detecting, body part location, age estimation and sex identification in the network picture, flow process such as Fig. 1.
Description of drawings
Fig. 1 is person detecting in the network picture, body part location, the age is estimated and the schematic flow sheet of the method and system of sex identification.
Embodiment
The purpose of this invention is to provide the method and system of person detecting in a kind of network picture, body part location, age estimation and sex identification, having comprised: one, to people's face sample training people face sorter; Two, to pedestrian's sample training pedestrian sorter; Three, to people's face sample training AAM model data; Four, to people's face sample training age in different sexes all ages and classes stage and the sorter of sex identification; Five, load sorter separately, be implemented in person detecting, body part location, age estimation and sex identification in the network picture; Concrete steps are as follows:
One, people's face sorter training process
Gather the positive negative sample of people's face, do histogram equalization and be normalized to 24 * 24 sizes;
To people's face sample extraction Haar-Like feature;
Import face characteristic into the Adaboost training, obtain people's face sorter;
Two, pedestrian's sorter training process
Gather the positive negative sample of pedestrian, do histogram equalization and be normalized to 64 * 128 sizes;
To pedestrian's sample extraction HOG feature;
Import pedestrian's sample characteristics into the SVM training, obtain pedestrian's sorter;
Three, AAM model data training process
Gather the positive samples pictures of people's face, wherein comprise each sex and each age level;
People's face samples pictures is gathered 68 unique points, and preserve its positional information;
Import the result of calculation of characteristic point position information and AAM into the SVM training, obtain the AAM model data;
Four, age, gender sorter were trained
Gather the positive samples pictures of people's face, wherein comprise each sex (being divided into man and woman) and each age level (being divided into teenage, adult and old), do histogram equalization and be normalized to 128 * 128 sizes;
To handling back sample extraction GIST feature;
Import sample characteristics into the SVM training, obtain age and gender sorter;
Five, spots localization and age-sex's identifying
The image network picture is carried out histogram equalization to be handled;
Add manned face sorter and pedestrian's sorter, detect human face region and human body zone;
If there is people's face, then load the AAM model data, face position, location; Load age-sex's sorter, judge age and sex;
How to exist the pedestrian then according to geometric relationship location parts of body, and do fusion with people's face positional information, make people's face produce corresponding with body position.
Claims (8)
1. the method and system that the purpose of this invention is to provide person detecting in a kind of network picture, body part location, age estimation and sex identification, comprise: to people's face sample training people face sorter, to pedestrian's sample training pedestrian sorter, to people's face sample training AAM model data, to people's face sample training age in different sexes all ages and classes stage and the sorter of sex identification, load sorter separately then, be implemented in person detecting, body part location, age estimation and sex identification in the network picture.
2. the sorter training process comprises: by gathering people's face, pedestrian's plus or minus samples pictures, do histogram equalization and be normalized to a certain size, extract corresponding sample characteristics, and import sample characteristics into the SVM training, thereby obtain corresponding sorter.
3. spots localization and age-sex's identifying comprise: the image network picture is carried out histogram equalization handle; Add manned face sorter and pedestrian's sorter, detect human face region and human body zone; If there is corresponding identifying object in the picture, then load corresponding sorter, thus face position, location; Load age-sex's sorter, judge age and sex; How to exist the pedestrian then according to geometric relationship location parts of body, and do fusion with people's face positional information, make people's face produce corresponding with body position.
4. people's face sorter training process as claimed in claim 1 is characterized in that: by gathering the positive negative sample of people's face, do histogram equalization and be normalized to 24 * 24 sizes; To people's face sample extraction Haar-Like feature; Import face characteristic into the Adaboost training, obtain people's face sorter.
5. pedestrian's sorter training process as claimed in claim 2 is characterized in that: by gathering the positive negative sample of pedestrian, do histogram equalization and be normalized to 64 * 128 sizes; To pedestrian's sample extraction HOG feature; Import pedestrian's sample characteristics into the SVM training, obtain pedestrian's sorter.
6. AAM model data training process as claimed in claim 3 is characterized in that: by gathering the positive samples pictures of people's face, wherein comprise each sex and each age level; People's face samples pictures is gathered 68 unique points, and preserve its positional information; Import the result of calculation of characteristic point position information and AAM into the SVM training, obtain the AAM model data.
7. age as claimed in claim 4, gender sorter training process, it is characterized in that by gathering the positive samples pictures of people's face, wherein comprise each sex (being divided into man and woman) and each age level (being divided into teenage, adult and old), do histogram equalization and be normalized to 128 * 128 sizes; To handling back sample extraction GIST feature; Import sample characteristics into the SVM training, obtain age and gender sorter.
8. spots localization as claimed in claim 5 and age-sex's identifying is characterized in that: handle by the image network picture being carried out histogram equalization; Add manned face sorter and pedestrian's sorter, detect human face region and human body zone; If there is people's face, then load the AAM model data, face position, location; Load age-sex's sorter, judge age and sex; How to exist the pedestrian then according to geometric relationship location parts of body, and do fusion with people's face positional information, make people's face produce corresponding with body position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100427898A CN103294982A (en) | 2012-02-24 | 2012-02-24 | Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100427898A CN103294982A (en) | 2012-02-24 | 2012-02-24 | Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103294982A true CN103294982A (en) | 2013-09-11 |
Family
ID=49095821
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012100427898A Pending CN103294982A (en) | 2012-02-24 | 2012-02-24 | Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103294982A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160317A (en) * | 2015-08-31 | 2015-12-16 | 电子科技大学 | Pedestrian gender identification method based on regional blocks |
| CN105303159A (en) * | 2015-09-17 | 2016-02-03 | 中国科学院合肥物质科学研究院 | Far-infrared pedestrian detection method based on distinctive features |
| CN106056083A (en) * | 2016-05-31 | 2016-10-26 | 腾讯科技（深圳）有限公司 | Information processing method and terminal |
| CN106815848A (en) * | 2017-01-17 | 2017-06-09 | 厦门可睿特信息科技有限公司 | Portrait background separation and contour extraction method based on grubcut and artificial intelligence |
| CN108509994A (en) * | 2018-03-30 | 2018-09-07 | 百度在线网络技术（北京）有限公司 | character image clustering method and device |
| CN108932783A (en) * | 2018-09-19 | 2018-12-04 | 南京邮电大学 | A kind of access control system towards big flow scene based on two-dimension human face identification |
| CN109543546A (en) * | 2018-10-26 | 2019-03-29 | 复旦大学 | The gait age estimation method returned based on the distribution of depth sequence |
| CN109784240A (en) * | 2018-12-30 | 2019-05-21 | 深圳市明日实业有限责任公司 | A kind of character recognition method, device and storage device |
| CN109934149A (en) * | 2019-03-06 | 2019-06-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for output information |
| CN110210293A (en) * | 2018-10-30 | 2019-09-06 | 上海市服装研究所有限公司 | A kind of gender identification method based on three-dimensional data and face-image |
| CN110232309A (en) * | 2018-10-30 | 2019-09-13 | 上海市服装研究所有限公司 | A method of based on three-dimensional data and neural network recognization sex character |
| WO2019227720A1 (en) * | 2018-06-01 | 2019-12-05 | 平安科技（深圳）有限公司 | Person image identification method, server and computer-readable storage medium |
| CN110852814A (en) * | 2020-01-14 | 2020-02-28 | 深圳惠通天下信息技术有限公司 | Advertisement delivery self-service system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101499018A (en) * | 2008-01-28 | 2009-08-05 | 联想(北京)有限公司 | Data processing unit and method |
| CN101950358A (en) * | 2010-09-30 | 2011-01-19 | 冠捷显示科技（厦门）有限公司 | Method for automatically estimating age and judging sex by intelligent television |
| US20110222724A1 (en) * | 2010-03-15 | 2011-09-15 | Nec Laboratories America, Inc. | Systems and methods for determining personal characteristics |
-
2012
- 2012-02-24 CN CN2012100427898A patent/CN103294982A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101499018A (en) * | 2008-01-28 | 2009-08-05 | 联想(北京)有限公司 | Data processing unit and method |
| US20110222724A1 (en) * | 2010-03-15 | 2011-09-15 | Nec Laboratories America, Inc. | Systems and methods for determining personal characteristics |
| CN101950358A (en) * | 2010-09-30 | 2011-01-19 | 冠捷显示科技（厦门）有限公司 | Method for automatically estimating age and judging sex by intelligent television |
Non-Patent Citations (1)
| Title |
|---|
| 陆丽: ""基于人脸图像的性别识别与年龄估计研究"", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160317A (en) * | 2015-08-31 | 2015-12-16 | 电子科技大学 | Pedestrian gender identification method based on regional blocks |
| CN105160317B (en) * | 2015-08-31 | 2019-02-15 | 电子科技大学 | One kind being based on area dividing pedestrian gender identification method |
| CN105303159A (en) * | 2015-09-17 | 2016-02-03 | 中国科学院合肥物质科学研究院 | Far-infrared pedestrian detection method based on distinctive features |
| CN106056083A (en) * | 2016-05-31 | 2016-10-26 | 腾讯科技（深圳）有限公司 | Information processing method and terminal |
| CN106056083B (en) * | 2016-05-31 | 2019-08-13 | 腾讯科技（深圳）有限公司 | A kind of information processing method and terminal |
| CN106815848A (en) * | 2017-01-17 | 2017-06-09 | 厦门可睿特信息科技有限公司 | Portrait background separation and contour extraction method based on grubcut and artificial intelligence |
| CN108509994A (en) * | 2018-03-30 | 2018-09-07 | 百度在线网络技术（北京）有限公司 | character image clustering method and device |
| WO2019227720A1 (en) * | 2018-06-01 | 2019-12-05 | 平安科技（深圳）有限公司 | Person image identification method, server and computer-readable storage medium |
| CN108932783A (en) * | 2018-09-19 | 2018-12-04 | 南京邮电大学 | A kind of access control system towards big flow scene based on two-dimension human face identification |
| CN109543546A (en) * | 2018-10-26 | 2019-03-29 | 复旦大学 | The gait age estimation method returned based on the distribution of depth sequence |
| CN109543546B (en) * | 2018-10-26 | 2022-12-20 | 复旦大学 | Gait age estimation method based on depth sequence distribution regression |
| CN110232309A (en) * | 2018-10-30 | 2019-09-13 | 上海市服装研究所有限公司 | A method of based on three-dimensional data and neural network recognization sex character |
| CN110210293A (en) * | 2018-10-30 | 2019-09-06 | 上海市服装研究所有限公司 | A kind of gender identification method based on three-dimensional data and face-image |
| CN109784240A (en) * | 2018-12-30 | 2019-05-21 | 深圳市明日实业有限责任公司 | A kind of character recognition method, device and storage device |
| CN109784240B (en) * | 2018-12-30 | 2023-08-22 | 深圳市明日实业有限责任公司 | Character recognition method, device and storage device |
| CN109934149A (en) * | 2019-03-06 | 2019-06-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for output information |
| CN109934149B (en) * | 2019-03-06 | 2022-08-09 | 百度在线网络技术（北京）有限公司 | Method and apparatus for outputting information |
| CN110852814A (en) * | 2020-01-14 | 2020-02-28 | 深圳惠通天下信息技术有限公司 | Advertisement delivery self-service system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103294982A (en) | Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network | |
| CN101584575B (en) | Age assessment method based on face recognition technology | |
| WO2015165365A1 (en) | Facial recognition method and system | |
| WO2019090769A1 (en) | Human face shape recognition method and apparatus, and intelligent terminal | |
| CN105160317B (en) | One kind being based on area dividing pedestrian gender identification method | |
| US9224037B2 (en) | Apparatus and method for controlling presentation of information toward human object | |
| CN108062546B (en) | Computer face emotion recognition system | |
| CN105787442B (en) | A kind of wearable auxiliary system and its application method of the view-based access control model interaction towards disturbance people | |
| WO2019033571A1 (en) | Facial feature point detection method, apparatus and storage medium | |
| CN104143079A (en) | Method and system for face attribute recognition | |
| CN104881660A (en) | Facial expression recognition and interaction method based on GPU acceleration | |
| CN105740780A (en) | Method and device for human face in-vivo detection | |
| CN104143076A (en) | Matching method and system for face shape | |
| CN102184016B (en) | Noncontact type mouse control method based on video sequence recognition | |
| CN105320948A (en) | Image based gender identification method, apparatus and system | |
| CN110728242A (en) | Image matching method and device based on portrait recognition, storage medium and application | |
| CN101719223A (en) | Identification method for stranger facial expression in static image | |
| WO2015131571A1 (en) | Method and terminal for implementing image sequencing | |
| WO2019212501A1 (en) | Trained recognition models | |
| CN116266415A (en) | Action evaluation method, system and device based on body building teaching training and medium | |
| Javed et al. | An intelligent alarm based visual eye tracking algorithm for cheating free examination system | |
| KR101787255B1 (en) | Facial expression recognition method based on ratio of facial ladnmark's distance | |
| WO2021235440A1 (en) | Method and device for acquiring movement feature amount using skin information | |
| Deák et al. | Estimation of human body height from uncalibrated image | |
| CN109901701A (en) | A kind of mouse action system based on computer vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130911 |