CN107145863A - A kind of Face detection method and system - Google Patents
A kind of Face detection method and system Download PDFInfo
- Publication number
- CN107145863A CN107145863A CN201710316792.7A CN201710316792A CN107145863A CN 107145863 A CN107145863 A CN 107145863A CN 201710316792 A CN201710316792 A CN 201710316792A CN 107145863 A CN107145863 A CN 107145863A
- Authority
- CN
- China
- Prior art keywords
- face
- location
- facial image
- camera
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of Face detection method and system, this method includes:The facial image of student is shot by the camera on desk;Detect the face location in facial image;After face location is detected, the status information of face is obtained according to default face Critical point model, determines that face key point this method realizes the accuracy for improving Face detection.
Description
Technical field
The present invention relates to technical field of face recognition, more particularly to a kind of Face detection method and system.
Background technology
At present, face recognition technology is a very active research field, and it covers Digital Image Processing, pattern and known
Not, the content of the subjects such as computer vision, neutral net, psychology, physiology, mathematics.With computer technology and light
Technology is continued to develop, and face recognition technology is more and more applied in the work and life of people.For example, face is known
Other attendance checking system, face recognition door control system, recognition of face shooting certificates handling system, and load on numerous APP of electronic product,
Such as U.S. elegant and the U.S. face camera of figure, even brush face are opened an account.Although the application of face recognition technology is widely how
Face is accurately positioned, the precision and the degree of accuracy for improving recognition of face are still that face recognition technology needs what is urgently solved to ask
Topic.
During recognition of face, the recognition of face especially to child, due to the characteristic of child's sports-like, key point during identification
Relative difficulty is positioned, and the distance of identification can not be fixed, the influence of picture pick-up device itself resolution factor in addition so as to child
The identification positioning relative difficulty of face, therefore the accuracy of identification positioning is relatively low.
The content of the invention
It is an object of the invention to provide a kind of Face detection method and system, to realize the accuracy for improving Face detection.
In order to solve the above technical problems, the present invention provides a kind of Face detection method, including:
The facial image of student is shot by the camera on desk;
Detect the face location in facial image;After face location is detected, according to default face Critical point model
The status information of face is obtained, face key point is determined.
It is preferred that, the face location in the detection facial image, including:
The first frame facial image is obtained, the feature templates for importing face are subjected to full figure search, multiple images feature is generated;
Strong classifier is generated according to characteristics of image, using returning cascade compensation pattern search to face location.
It is preferred that, before the face location in the detection facial image, in addition to:
Space geometry distance between numerology stranger face and camera, utilization space geometric distance calculates the chi of wave filter
Degree.
It is preferred that, before the status information that face is obtained according to default face Critical point model, in addition to:
The face location detected is preserved, the initialization of the integral position information of face as the crucial point location of face is believed
Breath.
The present invention also provides a kind of Face detection system, for realizing the above method, including:
Camera, the facial image for shooting student;
Data processing centre, for detecting the face location in facial image;After face location is detected, according to default
Face Critical point model obtain face status information, determine face key point.
It is preferred that, data processing centre includes:
The feature templates for importing face, for obtaining the first frame facial image, are carried out full figure search, generation by search module
Multiple images feature;Strong classifier is generated according to characteristics of image, using returning cascade compensation pattern search to face location;
Locating module, for after face location is detected, the shape of face to be obtained according to default face Critical point model
State information, determines face key point.
It is preferred that, the data processing centre also includes:
Computing module, for the space geometry distance between numerology stranger face and camera, utilization space geometric distance
Calculate the yardstick of wave filter.
It is preferred that, the data processing centre also includes:
Preserving module, for preserving the face location detected, regard the integral position information of face as face key point
The initialization information of positioning.
A kind of Face detection method and system provided by the present invention, the face of student is shot by the camera on desk
Image;Detect the face location in facial image;After face location is detected, obtained according to default face Critical point model
The status information of face, determines face key point.It can be seen that, the face information of student is shot by the camera on desk, and examine
Face location in altimetric image, after face location is detected, completes fixed according to the face Critical point model pre-defined
Position, the status information of real-time estimation face determines face key point by status information, improves the accuracy of Face detection.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is a kind of flow chart of Face detection method provided by the present invention;
Fig. 2 is children's Face detection parser outline flowchart;
Fig. 3 is a kind of structural representation of Face detection system provided by the present invention.
Embodiment
The core of the present invention is to provide a kind of Face detection method and system, to realize the accuracy for improving Face detection.
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only
A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is refer to, Fig. 1 is a kind of flow chart of Face detection method provided by the present invention, and this method includes:
S11:The facial image of student is shot by the camera on desk;
S12:Detect the face location in facial image;After face location is detected, according to default face key point
Model obtains the status information of face, determines face key point.
It can be seen that, this method shoots the face information of student, and the face position in detection image by the camera on desk
Put, after face location is detected, completed to position according to the face Critical point model pre-defined, real-time estimation face
Status information, face key point is determined by status information, improves the accuracy of Face detection.
Based on the above method, further, in step S12, the process of the face location in detection facial image is specially:
The first frame facial image is obtained, the feature templates for importing face are subjected to full figure search, multiple images feature is generated;According to image
Feature generates strong classifier, using returning cascade compensation pattern search to face location.
Further, in step S12, before the face location in detection facial image, in addition to:Numerology stranger face with
Space geometry distance between camera, utilization space geometric distance calculates the yardstick of wave filter.
Further, in step S12, before the status information that face is obtained according to default face Critical point model, also
Including:Preserve the face location that detects, using the integral position information of face as the crucial point location of face initialization information.
Detailed, this method is exactly a kind of Face detection algorithm, after classroom instruction starts, and Face detection algorithm is taught with course
Case synchronous averaging, then shoots the face information of student, and the face location in detection image by the camera on desk.When
Detect after face location, algorithm completes to position according to the face Critical point model pre-defined, real-time estimation face
Status information, completes to be accurately positioned child's face by the change of face key point.
Fig. 2 is children's Face detection parser outline flowchart, and based on this method, specific implementation process is as follows:
Step 1:The camera being arranged on desk starts simultaneously with course intelligence PPT, and will collect face information, regard
Frequency information transmission is to information processing centre;
Step 2:Information processing centre utilize according to the position of Intelligent seat calculate the space geometry of face and camera away from
From scaling filter is realized in design.According to the face length L of 3-6 Sui children, the dimension calculation of wave filter is:
Wherein, f is the focal length of camera, and D is distance of the camera to child position;
Step 3:When the first two field picture is reached, the feature templates (four hAAR features) for importing face are subjected to full figure and searched
Rope;Wherein, subtracting each other the gray value in the gray value and white matrix in grey matrix in template, generates substantial amounts of image special
Levy;
Step 4:In the feature of generation, each feature is done and trained on a Weak Classifier, all positive negative samples, finds one
Individual threshold value makes its classification error rate minimum, so as to generate strong classifier;
Step 5:Using returning the position that cascade compensation pattern search occurs to face in video image;Wherein, it is necessary to
Strong classifier that will be all detects face in same position, and algorithm just determines the position of child's face, and will be observed that
Face characteristic registration storage, and will generate face template is that face characteristic file is saved in database and forms feature templates;
Step 6:The face location detected is preserved, and regard the integral position information of face as the initial of crucial point location
Change information;
Step 7:When next two field picture is reached, positive negative sample, and profit will be gathered using the circular matrix of target peripheral region
With ridge regression training objective detector;
Step 8:Further, the ridge regression of linear space is mapped to non-linear space by kernel function, non-linear
Space calculates completion face tracking by solving dual problem, and simplifying using the diagonalization of circular matrix Fourier space.
Scaling filter herein is combined with the platform of intelligent desk, completes being accurately positioned for face.This method is first
Face tracking positioning is completed by returning cascade compensation model inspection face, and combining the method for correlation filtering.Child enters first
Enter classroom, as PPT broadcasting can attract sitting in the specified location of intelligent desk with merging for child's active, class
Study also starts immediately, and now the camera on intelligent desk has been switched on.Background information processing center is caught by camera
Realtime graphic carry out face collection, detection, position.The present invention can also increase after the crucial point location such as eyes, eyebrow, face
Concentration and emotional change to child are analyzed, so as to help child to improve learning efficiency, and help teacher and parent
Understand and correct in time the unhealthy emotion of child, form good personality and conduct.
Wherein, filtering refer to the operation that specific band frequency is filtered out in signal, be suppress and prevent interference one it is important
Measure, is the result of a certain random process according to the observation, the probability theory estimated another random process associated therewith
With method.
Fig. 3 is refer to, Fig. 3 is a kind of structural representation of Face detection system provided by the present invention, and the system is used for
The above method is realized, the system includes:
Camera 101, the facial image for shooting student;
Data processing centre 102, for detecting the face location in facial image;After face location is detected, foundation
Default face Critical point model obtains the status information of face, determines face key point.
It can be seen that, the system shoots the face information of student, and the face position in detection image by the camera on desk
Put, after face location is detected, completed to position according to the face Critical point model pre-defined, real-time estimation face
Status information, face key point is determined by status information, improves the accuracy of Face detection.
Based on said system, further, data processing centre includes:
The feature templates for importing face are carried out full figure search by search module specifically for obtaining the first frame facial image,
Generate multiple images feature;Strong classifier is generated according to characteristics of image, using returning cascade compensation pattern search to face location;
Locating module, for after face location is detected, the shape of face to be obtained according to default face Critical point model
State information, determines face key point.
Further, data processing centre also includes:Computing module, for the sky between numerology stranger face and camera
Between geometric distance, utilization space geometric distance calculate wave filter yardstick.
Further, data processing centre also includes:Preserving module, for preserving the face location detected, by face
Integral position information as the crucial point location of face initialization information.
The embodiment of each in this specification is described by the way of progressive, and what each embodiment was stressed is and other
Between the difference of embodiment, each embodiment identical similar portion mutually referring to.For system disclosed in embodiment
For, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is said referring to method part
It is bright.
A kind of Face detection method and system provided by the present invention are described in detail above.It is used herein
Specific case is set forth to the principle and embodiment of the present invention, and the explanation of above example is only intended to help and understands this
The method and its core concept of invention.It should be pointed out that for those skilled in the art, not departing from this hair
On the premise of bright principle, some improvement and modification can also be carried out to the present invention, these are improved and modification also falls into power of the present invention
In the protection domain that profit is required.
Claims (8)
1. a kind of Face detection method, it is characterised in that including:
The facial image of student is shot by the camera on desk;
Detect the face location in facial image;After face location is detected, obtained according to default face Critical point model
The status information of face, determines face key point.
2. the method as described in claim 1, it is characterised in that the face location in the detection facial image, including:
The first frame facial image is obtained, the feature templates for importing face are subjected to full figure search, multiple images feature is generated;Foundation
Characteristics of image generates strong classifier, using returning cascade compensation pattern search to face location.
3. method as claimed in claim 2, it is characterised in that before the face location in the detection facial image, also wrap
Include:
Space geometry distance between numerology stranger face and camera, utilization space geometric distance calculates the yardstick of wave filter.
4. method as claimed in claim 3, it is characterised in that described to obtain face according to default face Critical point model
Before status information, in addition to:
Preserve the face location that detects, using the integral position information of face as the crucial point location of face initialization information.
5. a kind of Face detection system, it is characterised in that for realizing the method as described in any one in Claims 1-4,
Including:
Camera, the facial image for shooting student;
Data processing centre, for detecting the face location in facial image;After face location is detected, according to default people
Face Critical point model obtains the status information of face, determines face key point.
6. system as claimed in claim 5, it is characterised in that data processing centre includes:
The feature templates for importing face, for obtaining the first frame facial image, are carried out full figure search, generated multiple by search module
Characteristics of image;Strong classifier is generated according to characteristics of image, using returning cascade compensation pattern search to face location;
Locating module, for after face location is detected, the state letter of face to be obtained according to default face Critical point model
Breath, determines face key point.
7. system as claimed in claim 6, it is characterised in that the data processing centre also includes:
Computing module, for the space geometry distance between numerology stranger face and camera, utilization space geometric distance is calculated
The yardstick of wave filter.
8. system as claimed in claim 7, it is characterised in that the data processing centre also includes:
Preserving module, for preserving the face location detected, regard the integral position information of face as the crucial point location of face
Initialization information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710316792.7A CN107145863A (en) | 2017-05-08 | 2017-05-08 | A kind of Face detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710316792.7A CN107145863A (en) | 2017-05-08 | 2017-05-08 | A kind of Face detection method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107145863A true CN107145863A (en) | 2017-09-08 |
Family
ID=59777891
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710316792.7A Pending CN107145863A (en) | 2017-05-08 | 2017-05-08 | A kind of Face detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107145863A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108256499A (en) * | 2018-02-05 | 2018-07-06 | 湖南科乐坊教育科技股份有限公司 | A kind of book assessment information determines method, apparatus and system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101499128A (en) * | 2008-01-30 | 2009-08-05 | 中国科学院自动化研究所 | Three-dimensional human face action detecting and tracing method based on video stream |
CN105825189A (en) * | 2016-03-21 | 2016-08-03 | 浙江工商大学 | Device for automatically analyzing attendance rate and class concentration degree of college students |
CN106228293A (en) * | 2016-07-18 | 2016-12-14 | 重庆中科云丛科技有限公司 | teaching evaluation method and system |
CN106250822A (en) * | 2016-07-21 | 2016-12-21 | 苏州科大讯飞教育科技有限公司 | Student's focus based on recognition of face monitoring system and method |
CN106295511A (en) * | 2016-07-26 | 2017-01-04 | 北京小米移动软件有限公司 | Face tracking method and device |
-
2017
- 2017-05-08 CN CN201710316792.7A patent/CN107145863A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101499128A (en) * | 2008-01-30 | 2009-08-05 | 中国科学院自动化研究所 | Three-dimensional human face action detecting and tracing method based on video stream |
CN105825189A (en) * | 2016-03-21 | 2016-08-03 | 浙江工商大学 | Device for automatically analyzing attendance rate and class concentration degree of college students |
CN106228293A (en) * | 2016-07-18 | 2016-12-14 | 重庆中科云丛科技有限公司 | teaching evaluation method and system |
CN106250822A (en) * | 2016-07-21 | 2016-12-21 | 苏州科大讯飞教育科技有限公司 | Student's focus based on recognition of face monitoring system and method |
CN106295511A (en) * | 2016-07-26 | 2017-01-04 | 北京小米移动软件有限公司 | Face tracking method and device |
Non-Patent Citations (3)
Title |
---|
JOAO F. HENRIQUES: "High-Speed Tracking with Kernelized Correlation Filters", 《IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE》 * |
兰胜坤: "基于 Adaboost 算法的人脸检测分类器的设计与实现", 《电脑知识与技术》 * |
梁硕 等: "基于 KCF 框架的长时间视频目标跟踪算法", 《信息传输与接入技术》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108256499A (en) * | 2018-02-05 | 2018-07-06 | 湖南科乐坊教育科技股份有限公司 | A kind of book assessment information determines method, apparatus and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109522815B (en) | Concentration degree evaluation method and device and electronic equipment | |
US11849998B2 (en) | Method for pupil detection for cognitive monitoring, analysis, and biofeedback-based treatment and training | |
US8015132B2 (en) | System and method for object detection and classification with multiple threshold adaptive boosting | |
CN104143079B (en) | The method and system of face character identification | |
CN105160318B (en) | Lie detecting method based on facial expression and system | |
CN103632136B (en) | Human-eye positioning method and device | |
US9892315B2 (en) | Systems and methods for detection of behavior correlated with outside distractions in examinations | |
US20210004587A1 (en) | Image detection method, apparatus, device and storage medium | |
CN106709450A (en) | Recognition method and system for fingerprint images | |
CN105224285A (en) | Eyes open and-shut mode pick-up unit and method | |
CN103593648B (en) | Face recognition method for open environment | |
US20140247993A1 (en) | Landmark localization via visual search | |
CN110634116A (en) | Facial image scoring method and camera | |
CN105809713A (en) | Object tracing method based on online Fisher discrimination mechanism to enhance characteristic selection | |
Shrivastava et al. | Conceptual model for proficient automated attendance system based on face recognition and gender classification using Haar-Cascade, LBPH algorithm along with LDA model | |
CN102314598A (en) | Retinex theory-based method for detecting human eyes under complex illumination | |
Sakthimohan et al. | Detection and Recognition of Face Using Deep Learning | |
CN111582654B (en) | Service quality evaluation method and device based on deep cycle neural network | |
CN107145863A (en) | A kind of Face detection method and system | |
CN109658357A (en) | A kind of denoising method towards remote sensing satellite image | |
CN110287759A (en) | A kind of eye strain detection method based on simplified input convolutional neural networks O-CNN | |
Zhao et al. | Fast localization algorithm of eye centers based on improved hough transform | |
CN104598866A (en) | Face-based social intelligence promotion method and system | |
CN107317846A (en) | Vehicle management system based on cloud platform | |
CN104091160B (en) | A kind of method for detecting human face |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170908 |