CN109711392A - A kind of talent's assessment method based on recognition of face - Google Patents
A kind of talent's assessment method based on recognition of face Download PDFInfo
- Publication number
- CN109711392A CN109711392A CN201910069942.8A CN201910069942A CN109711392A CN 109711392 A CN109711392 A CN 109711392A CN 201910069942 A CN201910069942 A CN 201910069942A CN 109711392 A CN109711392 A CN 109711392A
- Authority
- CN
- China
- Prior art keywords
- sample
- class
- subgraph
- face
- facial 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000001815 facial effect Effects 0.000 claims abstract description 35
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 230000000295 complement effect Effects 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 abstract description 4
- 230000035945 sensitivity Effects 0.000 abstract description 3
- 210000001331 nose Anatomy 0.000 description 9
- 210000000537 nasal bone Anatomy 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000003796 beauty Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The present invention provides a kind of talent's assessment method based on recognition of face, this method passes through acquisition facial image, and default class is carried out to facial image and divides to form subgraph, after to subgraph, general image feature extraction, it can accomplish that same class characteristic value matches, and then merge the object matching degree result for being formed and being completed.Method provided by the present invention, specific using fusion dividing sub-picture, the face identification method of multiclass SVM, high sensitivity, accuracy are high, calculating speed is fast, and can more adapt to the complex environment of face identification system realization.
Description
Technical field
The present invention relates to the Image mining technical field in big data, specially a kind of talent based on recognition of face
Assessment method.
Background technique
Talent assessment is the research achievement with pop psychology, management and related discipline, passes through psychological test, situation
The methods that objectify such as simulation measure the factors such as the ability, level, character trait of people, and according to position demand and enterprise's group
Knit characteristic to the psychological characteristics such as its quality situation, development potentiality, characteristics of personality make science evaluation, for enterprises recruit persons for jobs, choose,
It cultivates talent and human resource managements and development is waited to provide valuable reference information.Psychological test, interview and assessment centers are existing
For three kinds of main methods of talent assessment.Wherein the application of psychological test it is the most convenient with it is common.
But existing evaluation technology haves the defects that some to be difficult to overcome.For psychological test, be generally divided into " from
Comment " and " he comments " two ways: " self-appraisal " is easy to produce dummy results, and " he comments " is easy the shadow by subjective impression and life event
It rings；Therefore, there is a certain error for Psychological Evaluation, needs that other assessment methods is cooperated to be used in conjunction with.Assessment centers are a kind of new
The talent assessment tool of type is primarily referred to as the serial evaluation technology using Scene Simulation as core in the narrow sense；Assessment centers skill
Art is the synthesis of multi-method, more technologies, but it to the assessment time, place, topic setting, result evaluation etc. it is more demanding,
It is limited to be applicable in post, and is unsuitable for extensive testing.
For interview, that most of interviewer employs is the people that they like, rather than most go-getter, most of to determine
Plan person is just made that decision whether employing in initial 5 minutes of interview, and remaining time of interview is used to as theirs
Selection is justified oneself.That is in interview, estimator analyzes face, the speech of applicant according to itself experience and understanding
The information such as what is said or talked about, manner, and the beauty and ugliness quality of face phase influences a people and is self-confident or feels oneself inferior, and influences individual character, the temper of a people,
The glamour for influencing a people influences the ability and success rate of a people；This assessment method " is interviewed " from this view, essence
Upper is exactly a kind of " practising physiognomy ".
But there is also some distinctive defects for this measures: first is that observer is a kind of rough to the person of being observed
Rough assessment, even the physiognomy man of profession can only also make qualitative description, and the analysis of non-quantitation；Second is that observer
Assessment be only the person of being observed is evaluated according to experience, not only referring to information content and sample size it is limited, also deposit
In apparent Temporal change.
The content of present invention
In order to solve the above-mentioned technical problems, the present invention provides a kind of talent's assessment method based on recognition of face.
The present invention specifically adopts the following technical scheme that
A kind of talent's assessment method based on recognition of face, which comprises the following steps:
S01: the sample facial image of acquisition every profession and trade sample personnel, and store into sample database；
S02: carrying out default class according to face characteristic to the sample facial image and be divided into sample subgraph, and by the sample
Subgraph, the sample facial image classification storage are into sample storage；Wherein, the sample subgraph complementary overhangs；
S03: acquisition target facial image carries out default class according to the face characteristic to the target facial image and is divided into mesh
Subgraph is marked, and by the target subgraph, the target facial image classification storage into local storage；Wherein, described
Target subgraph complementary overhangs；
S04: feature extraction is carried out for the of a sort sample subgraph, the target subgraph respectively, respectively obtains sample
Eigen value, object feature value, and class match decision is carried out to the sample characteristics and the object feature value；
S05: repeating step S04 until all default class match decisions finish；It merges the class match decision to form the mesh
Mark the object matching degree result of facial image；
S06: the object matching degree is fed back as a result, generating the test and evaluation report of corresponding demand.
Preferably, in the present invention, in the step S02, the default class is face's global feature class, eye class, nose
Subclass, mouth class, cheek class.
Preferably, face's entirety class includes profile, shape, size, relative position；The eye class includes eyes
Position, eye exterior feature, eyes size；The nose class includes nose shape, nasal bone height, nasal bone length；The mouth class includes mouth position
It sets, mouth size；The cheek class includes cheek shape, cheek size.
Preferably, in the present invention, in the step S04, the feature extraction is specially to use broad sense two-dimentional
Fisher linear discriminant analysis carries out local shape factor, global characteristics for above-mentioned sample subgraph, the target subgraph
It extracts.
Preferably, in the present invention, in the step S04, using Multi- class SVM classifier as each subgraph with it is whole
The classifier of body image.
Preferably, in the present invention, also have in step S07, it can timing for the sample database of different industries
It is updated.
Compared with prior art, the medicine have the advantages that
1. the present invention is passed through according to each professionalism interior on a large scale by the facial image characteristic information acquisition technique of intelligence
The channels such as acquisition face to face, network acquisition, obtain complete face database, and constantly update；
2. the present invention is by way of image procossing, specific using fusion dividing sub-picture, the face identification method of multiclass SVM, spirit
Sensitivity is high, accuracy is high, calculating speed is fast, and can more adapt to the complex environment of face identification system realization.
Detailed description of the invention
Fig. 1 is the method flow diagram of one embodiment of the invention.
Fig. 2 is the image processing flow figure in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1
As shown in Figure 1, the present invention provides a kind of talent's assessment method based on recognition of face, comprising the following steps:
S01: the sample facial image of acquisition every profession and trade sample personnel, and store into sample database；
S02: carrying out default class according to face characteristic to the sample facial image and be divided into sample subgraph, and by the sample
Subgraph, the sample facial image classification storage are into sample storage；Wherein, the sample subgraph complementary overhangs；
The default class is face's global feature class, eye class, nose class, mouth class, cheek class；Wherein, face's entirety class packet
Include profile, shape, size, relative position；The eye class includes eye position, eye exterior feature, eyes size；The nose class packet
Include nose shape, nasal bone height, nasal bone length；The mouth class includes mouth position, mouth size；The cheek class includes cheek shape
Shape, cheek size.
S03: acquisition target facial image carries out default class segmentation according to the face characteristic to the target facial image
At target subgraph, and by the target subgraph, the target facial image classification storage into local storage；Wherein,
The target subgraph complementary overhangs；
S04: feature extraction is carried out for the of a sort sample subgraph, the target subgraph respectively, respectively obtains sample
Eigen value, object feature value, and class match decision is carried out to the sample characteristics and the object feature value；
The feature extraction is specially to use broad sense two dimension Fisher linear discriminant analysis for above-mentioned sample subgraph, the mesh
It marks subgraph and carries out local shape factor, global characteristics extraction.
S05: repeating step S04 until all default class match decisions finish；It merges the class match decision to form institute
State the object matching degree result of target facial image；
Classifier using Multi- class SVM classifier as each subgraph and general image.
S06: the object matching degree is fed back as a result, generating the test and evaluation report of corresponding demand.
S07: the sample database of different industries can be periodically updated.
Embodiment 2
As shown in Figure 1, the present invention provides a kind of talent's assessment method based on recognition of face, comprising the following steps:
S01: the sample facial image of acquisition every profession and trade sample personnel, and store into sample database；
S02: carrying out default class according to face characteristic to the sample facial image and be divided into sample subgraph, and by the sample
Subgraph, the sample facial image classification storage are into sample storage；Wherein, the sample subgraph complementary overhangs；
The default class is face's global feature class, eye class, nose class, mouth class, cheek class；Wherein, face's entirety class packet
Include profile, shape, size, relative position；The eye class includes eye position, eye exterior feature, eyes size；The nose class packet
Include nose shape, nasal bone height, nasal bone length；The mouth class includes mouth position, mouth size；The cheek class includes cheek shape
Shape, cheek size.
S03: acquisition target facial image carries out default class segmentation according to the face characteristic to the target facial image
At target subgraph, and by the target subgraph, the target facial image classification storage into local storage；Wherein,
The target subgraph complementary overhangs；
S04: feature extraction is carried out for the of a sort sample subgraph, the target subgraph respectively, respectively obtains sample
Eigen value, object feature value, and class match decision is carried out to the sample characteristics and the object feature value；
The feature extraction is specially to use broad sense two dimension Fisher linear discriminant analysis for above-mentioned sample subgraph, the mesh
It marks subgraph and carries out local shape factor, global characteristics extraction.
S05: repeating step S04 until all default class match decisions finish；It merges the class match decision to form institute
State the object matching degree result of target facial image；
Classifier using Multi- class SVM classifier as each subgraph and general image.
S06: the object matching degree is fed back as a result, generating the test and evaluation report of corresponding demand.
S07: the sample database of different industries can be periodically updated.
As shown in Fig. 2, in the present invention, specifically handling in the following way acquired image:
A. general image, including sample facial image, target facial image are acquired；
B. subgraph is cut by default class to general image；Subgraph includes sample subgraph, target subgraph；
C. local feature value, the overall situation are extracted to general image, each subgraph using broad sense two dimension Fisher linear discriminant analysis
Characteristic value, the characteristic value include sample characteristic value, object feature value；
D. classifier of the multiclass SVM as each subgraph and general image is used, the class match decision fusion of each SVM is merged
And export object matching degree result.
Here, these belong to this field for how specifically using cutting formula, Feature Sensitivity, specificity parameter
The conventional means that technical staff can choose.
In order to accelerate to execute the speed of cluster, which can also include the steps that initial segmentation: will be whole
Image initial is divided into multiple images block, and each image block is the image of different parts, including multiple pixels；Wherein, for same
A kind of image carries out feature extraction, the characteristic value extracted is matched, to substantially increase extraction efficiency, precision
With matching efficiency；On this basis, repeat same class image characteristics extraction, until all class image characteristics extractions finish,
Fusion matching is carried out on this basis；To having reached the matched effect of whole strategy, enable method provided by the present invention
The complex environment that face identification system is realized more is adapted to, for example image data is big, fogging image or image personage have screening
Gear etc..In the present invention, the image partition method of use considers the Space Consistency of facial image, i.e. neighbouring facial image block
It is more likely to belong to this rule of same cut zone, to realize more acurrate and quick image segmentation.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (5)
1. a kind of talent's assessment method based on recognition of face, which is characterized in that talent's assessment method includes following step
It is rapid:
S01: the sample facial image of acquisition every profession and trade sample personnel, and store into sample database；
S02: carrying out default class according to face characteristic to the sample facial image and be divided into sample subgraph, and by the sample
Subgraph, the sample facial image classification storage are into sample storage；Wherein, the sample subgraph complementary overhangs；
S03: acquisition target facial image carries out default class according to the face characteristic to the target facial image and is divided into mesh
Subgraph is marked, and by the target subgraph, the target facial image classification storage into local storage；Wherein, described
Target subgraph complementary overhangs；
S04: feature extraction is carried out for the of a sort sample subgraph, the target subgraph respectively, respectively obtains sample
Eigen value, object feature value, and class match decision is carried out to the sample characteristics and the object feature value；
S05: repeating step S04 until all default class match decisions finish；It merges the class match decision to form the mesh
Mark the object matching degree result of facial image；
S06: the object matching degree is fed back as a result, generating the test and evaluation report of corresponding demand.
2. a kind of talent's assessment method based on recognition of face according to claim 1, it is characterised in that: in the step
In S02, the default class is face's global feature class, eye class, nose class, mouth class, cheek class.
3. a kind of talent's assessment method based on recognition of face according to claim 1 or claim 2, it is characterised in that: in the step
In rapid S04, the feature extraction is specially that broad sense two dimension Fisher linear discriminant analysis is used to extract above-mentioned sample subgraph.
4. according to claim 1-3 any one of state a kind of talent's assessment method based on recognition of face, it is characterised in that: in institute
It states in step S04, the classifier using Multi- class SVM classifier as each subgraph and general image.
5. -4 any a kind of talent's assessment method based on recognition of face according to claim 1, it is characterised in that: also have
In step S07, the sample database of different industries can be periodically updated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910069942.8A CN109711392A (en) | 2019-01-24 | 2019-01-24 | A kind of talent's assessment method based on recognition of face |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910069942.8A CN109711392A (en) | 2019-01-24 | 2019-01-24 | A kind of talent's assessment method based on recognition of face |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109711392A true CN109711392A (en) | 2019-05-03 |
Family
ID=66262812
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910069942.8A Pending CN109711392A (en) | 2019-01-24 | 2019-01-24 | A kind of talent's assessment method based on recognition of face |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109711392A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1975759A (en) * | 2006-12-15 | 2007-06-06 | 中山大学 | Human face identifying method based on structural principal element analysis |
| CN101599174A (en) * | 2009-08-13 | 2009-12-09 | 哈尔滨工业大学 | Method for outline extraction of level set medical ultrasonic image area based on edge and statistical nature |
| CN104143079A (en) * | 2013-05-10 | 2014-11-12 | 腾讯科技（深圳）有限公司 | Method and system for face attribute recognition |
| CN104298753A (en) * | 2014-10-17 | 2015-01-21 | 重庆市云日信息技术有限公司 | Personnel assessment method based on face image processing |
| CN107330359A (en) * | 2017-05-23 | 2017-11-07 | 深圳市深网视界科技有限公司 | A kind of method and apparatus of face contrast |
| CN107423707A (en) * | 2017-07-25 | 2017-12-01 | 深圳帕罗人工智能科技有限公司 | A kind of face Emotion identification method based under complex environment |
| CN107516094A (en) * | 2017-10-12 | 2017-12-26 | 北京军秀咨询有限公司 | A kind of personal assessment methods based on face image processing |
| WO2018001092A1 (en) * | 2016-06-29 | 2018-01-04 | 中兴通讯股份有限公司 | Face recognition method and apparatus |
-
2019
- 2019-01-24 CN CN201910069942.8A patent/CN109711392A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1975759A (en) * | 2006-12-15 | 2007-06-06 | 中山大学 | Human face identifying method based on structural principal element analysis |
| CN101599174A (en) * | 2009-08-13 | 2009-12-09 | 哈尔滨工业大学 | Method for outline extraction of level set medical ultrasonic image area based on edge and statistical nature |
| CN104143079A (en) * | 2013-05-10 | 2014-11-12 | 腾讯科技（深圳）有限公司 | Method and system for face attribute recognition |
| CN104298753A (en) * | 2014-10-17 | 2015-01-21 | 重庆市云日信息技术有限公司 | Personnel assessment method based on face image processing |
| WO2018001092A1 (en) * | 2016-06-29 | 2018-01-04 | 中兴通讯股份有限公司 | Face recognition method and apparatus |
| CN107330359A (en) * | 2017-05-23 | 2017-11-07 | 深圳市深网视界科技有限公司 | A kind of method and apparatus of face contrast |
| CN107423707A (en) * | 2017-07-25 | 2017-12-01 | 深圳帕罗人工智能科技有限公司 | A kind of face Emotion identification method based under complex environment |
| CN107516094A (en) * | 2017-10-12 | 2017-12-26 | 北京军秀咨询有限公司 | A kind of personal assessment methods based on face image processing |
Non-Patent Citations (2)
| Title |
|---|
| 栗科峰 等: "基于子图分割与多类支持向量机的人脸识别方法", 《科技通报》 * |
| 栗科峰 等: "基于子图分割与多类支持向量机的人脸识别方法", 《科技通报》, vol. 34, no. 08, 31 August 2018 (2018-08-31), pages 1 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Herzlinger et al. | A new method for 3D geometric morphometric shape analysis: The case study of handaxe knapping skill | |
| Dehghan et al. | Dager: Deep age, gender and emotion recognition using convolutional neural network | |
| CN104866829B (en) | A kind of across age face verification method based on feature learning | |
| CN104298753B (en) | Personal assessment methods based on face image processing | |
| CN106777891B (en) | A kind of selection of data characteristics and prediction technique and device | |
| CN104239858B (en) | A kind of method and apparatus of face characteristic checking | |
| CN109711281A (en) | A kind of pedestrian based on deep learning identifies again identifies fusion method with feature | |
| CN109598186A (en) | A kind of pedestrian's attribute recognition approach based on multitask deep learning | |
| Kaya et al. | Multimodal personality trait analysis for explainable modeling of job interview decisions | |
| CN112101203A (en) | Facial beauty medical design system and method based on data analysis | |
| CN106055657B (en) | Viewing index assessment system for specific crowd | |
| CN109003259B (en) | Fingerprint quality evaluation method based on line quality expert visual cognition machine learning | |
| CN109711392A (en) | A kind of talent's assessment method based on recognition of face | |
| CN110660068B (en) | Semi-automatic brain region segmentation method for three-dimensional cell construction image | |
| CN110458751A (en) | A kind of face replacement method, equipment and medium based on Guangdong opera picture | |
| Undavia et al. | Customized prediction model to predict post-graduation course for graduating students using decision tree classifier | |
| JP7095849B1 (en) | Eyewear virtual fitting system, eyewear selection system, eyewear fitting system and eyewear classification system | |
| CN110298301A (en) | A kind of establishment officer's phychology prediction technique | |
| CN117152850A (en) | Customer service emotion video database construction based on multi-modal scoring and deep learning | |
| CN109614606B (en) | Document embedding-based long text case penalty range classification prediction method and device | |
| Bülthoff et al. | Personally familiar faces: Higher precision of memory for idiosyncratic than for categorical information. | |
| CN112818739A (en) | Image instrument dimension evaluation method and device for intelligent interview | |
| Gerlach et al. | Who loves virtue as much as he loves beauty?: Deep learning based estimator for aesthetics of portraits | |
| CN112836645A (en) | Large-scale exercise heart rate sequence-oriented running-instead detection method | |
| Deligio et al. | IT-and machine learning-based methods of classification: The cooperative project 2 ClaReNet–Classification and Representation 3 for Networks 4 |
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: 20190503 |