CN107239734A - A kind of three-dimensional face identification method for prison access management system - Google Patents
A kind of three-dimensional face identification method for prison access management system Download PDFInfo
- Publication number
- CN107239734A CN107239734A CN201710259447.4A CN201710259447A CN107239734A CN 107239734 A CN107239734 A CN 107239734A CN 201710259447 A CN201710259447 A CN 201710259447A CN 107239734 A CN107239734 A CN 107239734A
- Authority
- CN
- China
- Prior art keywords
- mrow
- dimensional
- human face
- histogram
- feature
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- 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
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
Abstract
The invention discloses a kind of three-dimensional face identification method for prison access management system, this method is used as tranining database using the three-dimensional face data of registered inmate and the personnel that visit a prisoner, using the registered object of Mesh LBP operator extractions and object to be detected local feature and be combined, obtain the global feature of description object.Then SVM models are trained using the three-dimensional face data of registered object, then object to be detected is put into SVM models and classified, the identity information of object according to belonging to being judged classification results.
Description
Technical field
The present invention relates to face identification method field, specifically a kind of three-dimensional face for prison access management system
Recognition methods.
Background technology
Prison is the organization for execution of punishment of country, is also the main place for putting in prison criminal.It is right due to the particularity of itself
Safety management has special requirement.Prison AB be into and out prison controlled area only way which must be passed, it is necessary to strictly supervised.Pass
The AB doors turnover of system is all visually to recognize the method being combined using IC-card and policeman on duty, and this method has certain disadvantage
Hold, the various factors such as operator on duty's night fatigue, operator on duty's sense of responsibility largely affects the safety of jails.Also,
Prison staff Migrant women is various, complicated component, and the high security that traditional artificial management and control mode can not adapt to prison will
Ask.
In recent years, biological identification technology has obtained the approval of more and more people.Because everyone biological characteristic and other people
Compare, with constant over a period to come stability and uniqueness, be difficult to be forged and palmed off, so biological identification technology
It is safer, reliably, accurately.Particularly face characteristic, is most important one kind in biological characteristic, reflects many human individuals
Important information, be people distinguish Different Individual most directly distinguish object.Therefore the present invention is paid close attention to and recognition of face skill
Art access management systematic difference at the prison.
At present, the two-dimension human face identification technology based on image reaches its maturity, and has been obtained under certain constraints preferably
Recognition result.However, due to illumination, posture, cosmetic, expression, the change at age, these factors significantly reduce two-dimension human face
The performance of recognizer.By contrast, three-dimensional face identification refers to face's 3D shape by the object to be identified obtained is gathered
Data are matched with face's three-dimensional shape data of difficulties that one is reluctant to bring to the notice of others known identities as identification one, then draw object to be identified
The process of identity.Three-dimensional face identification uses three-dimensional face data, and the advantage of three-dimensional face data is that holding
The original geometry information of face, is not influenceed by the factor such as illumination and expression, and three-dimensional data has than 2-D data (figure
Picture) more information content.However, at present on how directly to obtain reliable and succinct from the three-dimensional face data collected
Feature representation all the time without suitable achievement, and the achievement be used for three-dimensional face practice also lack practice, and this
A little accuracys rate for improving three-dimensional face identification are vital.
Therefore, the feature representation for how extracting three-dimensional face is explored, and one kind is proposed based on the feature representation obtained
New three-dimensional face identification method, this for three-dimensional face recognition accuracy raising highly significant, and then improve prison go out
The recognition performance of entrance management system plays a significant role.
The content of the invention
It is an object of the invention to provide a kind of three-dimensional face identification method for prison access management system, to solve
Prior art two-dimension human face recognizer is due under the presence recognition accuracy of the changes such as illumination, posture, cosmetic, expression, age
The problem of drop.
In order to achieve the above object, the technical solution adopted in the present invention is：
A kind of three-dimensional face identification method for prison access management system, it is characterised in that：Utilize mesh-lbp
It is overall that the local feature of operator extraction three-dimensional human face scanning figure regional and being combined obtains description three-dimensional human face scanning figure
Characteristic vector, then carries out SVM with known three-dimensional face features and trains to obtain training pattern, then three-dimensional face to be detected is used
Training pattern is classified, and obtains testing result, its step includes：
(1), the three-dimensional human face scanning figure for choosing multiple known objects is used as tranining database；
(2) region division, is carried out to the three-dimensional human face scanning figure of each object in tranining database, then to each region
Carry out triangle gridding；
(3), using the every width three-dimensional human face scanning figure of Mesh-LBP operator extractions regional local feature histogram,
The local feature histogram of regional is saved as into characteristic vector respectively, local Nogata is constituted by the local feature in each region
Shown in figure statistical nature, wherein Mesh-LBP operators such as formula (1)：
In formula (1), k refers to the grid numbering that outer collarette is set up by Mesh-LBP operators rule；M is the net of outer collarette
Lattice sum；fcIt is central gridding；fkIt is the grid of outer collarette；H (x) is defined in the scalar function on grid；α (k) is weight letter
Number, different follows the example of corresponding to different variants；S (x) is a binaryzation function, and definition is as shown in formula (2)：
(4), the local feature histogram of the regional of every width three-dimensional human face scanning figure is attached, obtaining description should
The global feature histogram of width three-dimensional human face scanning figure, using the global feature histogram as the width face characteristic vector, from
And local histogram's statistical natures all in every width three-dimensional human face scanning figure is combined into a histogram statistical features conduct
The global feature of the width three-dimensional human face scanning figure；
(5), the global feature for the every width three-dimensional human face scanning figure for obtaining step (4) is trained using SVM respectively, is instructed
Practice model；
(6) the global feature vector of object three-dimensional human face scanning figure to be measured, is extracted using identical method, test specimens are used as
This；
(7) test sample, is put into classification in training pattern and obtains classification results, three-dimensional face is calculated according to classification results
The accuracy rate of identification.
A kind of described three-dimensional face identification method for prison access management system, it is characterised in that：With prison
Registered inmate and the personnel that visit a prisoner as known object, using the registered inmate of Mesh-LBP operator extractions and
Visit a prisoner personnel three-dimensional human face scanning figure local feature histogram, and local feature histogram is combined obtain one and can retouch
The characteristic vector of three-dimensional face global feature is stated, the vectorial three-dimensional face of this feature is recognized, and then original confirmation prison gateway
The identity information of personnel.
A series of three-dimensional human face scanning figure that the present invention chooses known objects constitutes database, and every width in database is scanned
Figure zoning, then triangulation is first carried out to each region, it is rear with Mesh-Local Binary Pattern (Mesh-
LBP) local feature of operator extraction three-dimensional human face scanning figure constitutes local histogram's statistical nature, will own in a width scanning figure
Histogram statistical features be combined into a histogram statistical features as the global feature of this width three-dimensional human face scanning figure.Then
It is trained using SVM and draws training pattern, three-dimensional face global feature to be detected is divided using the model trained
Class, draws classification results, so that for identification.
Mesh-LBP operators directly act on triangle gridding manifold surface, and Mesh-LBP operators are retouched as a kind of texture
Fine three-dimensional Local textural feature can be extracted by stating operator, can provide very strong resolution performance.But three-dimensional Local textural feature
It is different from the two-dimentional Local textural feature obtained using LBP operators.Three-dimensional Local textural feature refers to make body surface " rise
The repeatable model of the three-dimensional of wrinkle ", and two-dimentional Local textural feature refers to the luminosity information of object.
SVM, which is one, the learning model of supervision, commonly used to carry out pattern-recognition, classification and regression analysis, and it is learned
It is margin maximization to practise strategy.The present invention uses Mesh-LBP operators, is maintaining the validity of basic LBP operators and can expand
While malleability, also maintain linear computation complexity, then by using SVM classifier significantly improve calculating efficiency and
Accuracy rate.
Basis of the invention by being trained from the three-dimensional face data of registered inmate and the personnel that visit a prisoner
On, the three-dimensional face of personnel to be measured is detected, so that the identification for personnel identity to be measured.The access management systematic difference
Be conducive to the specification for passing in and out Prison staff effectively to manage, ensure the safety of jails.
Compared with the prior art, beneficial effects of the present invention are embodied in：
Mesh-LBP operators are used to describe three-dimensional face features there is provided a kind of new three-dimensional face identification method, from
And complete relatively reliable identification task.
Brief description of the drawings
Fig. 1 is the implementation process frame diagram of the present invention.
Fig. 2 be the present invention experiment in three-dimensional face carry out triangle gridding, arbitrarily collinear three points connection be three
Angle grid.
Fig. 3 be the present invention experiment in three-dimensional face regional utilize Mesh-LBP operator extractions local histogram.
Embodiment
As shown in figure 1, a kind of three-dimensional face identification method for prison access management system, comprises the following steps：
(1), 2500 three-dimensional human face scanning figures for choosing 100 objects are used as tranining database.
(2) zoning, is carried out to every width scanning figure of each object, after triangle gridding is being carried out to each region.
(3), using the every width scanning figure of Mesh-LBP operator extractions each region local feature histogram, preserved
It is characterized vector.Mesh-LBP operators are as follows：
In formula (1), k refers to the grid numbering that outer collarette is set up by Mesh-LBP operators rule；M is the net of outer collarette
Lattice sum；fcIt is central gridding；fkIt is the grid of outer collarette；H (x) is defined in the scalar function on grid, can use curvature,
Gaussian curvature, curvature, four kinds of functions of face normal angle；α (k) is weighting function, and different follows the example of corresponding to different variants；s(x)
It is a binaryzation function, definition is as shown in formula (2)：
(4), each region histogram of every width scanning figure is attached, so that it is whole to obtain description three-dimensional human face scanning figure
The characteristic vector of body.
(5), three-dimensional face features are trained using SVM, training pattern is obtained.
(6) the global feature vector of object three-dimensional human face scanning figure to be measured, is extracted using identical method, test specimens are used as
This.
(7), test sample is put into training pattern and classified, the accurate of three-dimensional face identification is calculated according to classification results
Rate.
In Fig. 2, (a) figure takes in face two and three points of nose as datum mark first, then on the basis of datum mark
Upper utilization mathematics geometric knowledge generates each summit of triangle gridding, and the plane for generating point is positioned on three-dimensional face by (b) figure
I.e. to having carried out triangle gridding, the local feature Nogata that (c) desires to make money or profit with the Mesh-LBP operator extractions face to three-dimensional face
Figure, then connects and composes the global feature histogram of the face, that is, has obtained the overall description of the face.Image data comes from
BU-3DFE data sets.
In Fig. 3, (a) figure is that, from same target but the different face of posture, (b) is to be carried respectively using Mesh-LBP operators
The face obtained is totally described.It is seen that, this two width face describes similar, therefore can determine whether as same object.Picture number
According to coming from BU-3DFE data sets.By taking two width different gestures of same target as an example, be utilized respectively Mesh-LBP operators to this two
Width three-dimensional face extract obtained overall description, then contrasts the two and totally describes, you can complete appointing for Classification and Identification
Business.
Claims (2)
1. a kind of three-dimensional face identification method for prison access management system, it is characterised in that：Calculated using mesh-lbp
Son extracts the local feature of three-dimensional human face scanning figure regional and is combined and obtains describing three-dimensional human face scanning figure entirety
Characteristic vector, then carries out SVM training with known three-dimensional face features and obtains training pattern, then three-dimensional face to be detected is instructed
Practice category of model, obtain testing result, its step includes：
(1), the three-dimensional human face scanning figure for choosing multiple known objects is used as tranining database；
(2) region division, is carried out to the three-dimensional human face scanning figure of each object in tranining database, then each region carried out
Triangle gridding；
(3), using the every width three-dimensional human face scanning figure of Mesh-LBP operator extractions regional local feature histogram, will be each
The local feature histogram in individual region saves as characteristic vector respectively, and local histogram's system is constituted by the local feature in each region
Feature is counted, wherein shown in Mesh-LBP operators such as formula (1)：
<mrow>
<mi>M</mi>
<mi>e</mi>
<mi>s</mi>
<mi>h</mi>
<mo>-</mo>
<mi>L</mi>
<mi>B</mi>
<mi>P</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>c</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>m</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>h</mi>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>k</mi>
</msub>
<mo>)</mo>
<mo>-</mo>
<mi>h</mi>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>c</mi>
</msub>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&CenterDot;</mo>
<mi>&alpha;</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
In formula (1), k refers to the grid numbering that outer collarette is set up by Mesh-LBP operators rule；M is that the grid of outer collarette is total
Number；fcIt is central gridding；fkIt is the grid of outer collarette；H (x) is defined in the scalar function on grid；α (k) is weighting function,
Different follows the example of corresponding to different variants；S (x) is a binaryzation function, and definition is as shown in formula (2)：
<mrow>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mi>x</mi>
<mo>&GreaterEqual;</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>0</mn>
<mo>,</mo>
<mi>x</mi>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
(4), the local feature histogram of the regional of every width three-dimensional human face scanning figure is attached, obtains describing this three
Tie up human face scanning figure global feature histogram, using the global feature histogram as the width face characteristic vector so that
All local histogram's statistical natures are combined into a histogram statistical features as the width in every width three-dimensional human face scanning figure
The global feature of three-dimensional human face scanning figure；
(5), the global feature for the every width three-dimensional human face scanning figure for obtaining step (4) is trained using SVM respectively, obtains training mould
Type；
(6) the global feature vector of object three-dimensional human face scanning figure to be measured, is extracted using identical method, test sample is used as；
(7) test sample, is put into classification in training pattern and obtains classification results, three-dimensional face identification is calculated according to classification results
Accuracy rate.
2. a kind of three-dimensional face identification method for prison access management system according to claim 1, its feature
It is：It is registered using Mesh-LBP operator extractions using the registered inmate in prison and the personnel that visit a prisoner as known object
Inmate and the personnel that visit a prisoner three-dimensional human face scanning figure local feature histogram, and by local feature histogram combine
The characteristic vector of three-dimensional face global feature can be described to one, the vectorial three-dimensional face of this feature is recognized, so it is original true
Recognize the identity information of prison gateway personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710259447.4A CN107239734A (en) | 2017-04-20 | 2017-04-20 | A kind of three-dimensional face identification method for prison access management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710259447.4A CN107239734A (en) | 2017-04-20 | 2017-04-20 | A kind of three-dimensional face identification method for prison access management system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107239734A true CN107239734A (en) | 2017-10-10 |
Family
ID=59983068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710259447.4A Pending CN107239734A (en) | 2017-04-20 | 2017-04-20 | A kind of three-dimensional face identification method for prison access management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107239734A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110069968A (en) * | 2018-01-22 | 2019-07-30 | 耐能有限公司 | Face recognition and face recognition method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799870A (en) * | 2012-07-13 | 2012-11-28 | 复旦大学 | Single-training sample face recognition method based on blocking consistency LBP (Local Binary Pattern) and sparse coding |
| CN102867350A (en) * | 2012-08-29 | 2013-01-09 | 广东铂亚信息技术股份有限公司 | Prison AB door safety management method based on face recognition technology |
| US8803879B1 (en) * | 2002-07-10 | 2014-08-12 | Nvidia Corporation | Omnidirectional shadow texture mapping |
| CN103996052A (en) * | 2014-05-12 | 2014-08-20 | 深圳市唯特视科技有限公司 | Three-dimensional face gender classification device and method based on three-dimensional point cloud |
| CN104268539A (en) * | 2014-10-17 | 2015-01-07 | 中国科学技术大学 | High-performance human face recognition method and system |
| CN104504410A (en) * | 2015-01-07 | 2015-04-08 | 深圳市唯特视科技有限公司 | Three-dimensional face recognition device and method based on three-dimensional point cloud |
| CN105654035A (en) * | 2015-12-21 | 2016-06-08 | 湖南拓视觉信息技术有限公司 | Three-dimensional face recognition method and data processing device applying three-dimensional face recognition method |
-
2017
- 2017-04-20 CN CN201710259447.4A patent/CN107239734A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8803879B1 (en) * | 2002-07-10 | 2014-08-12 | Nvidia Corporation | Omnidirectional shadow texture mapping |
| CN102799870A (en) * | 2012-07-13 | 2012-11-28 | 复旦大学 | Single-training sample face recognition method based on blocking consistency LBP (Local Binary Pattern) and sparse coding |
| CN102867350A (en) * | 2012-08-29 | 2013-01-09 | 广东铂亚信息技术股份有限公司 | Prison AB door safety management method based on face recognition technology |
| CN103996052A (en) * | 2014-05-12 | 2014-08-20 | 深圳市唯特视科技有限公司 | Three-dimensional face gender classification device and method based on three-dimensional point cloud |
| CN104268539A (en) * | 2014-10-17 | 2015-01-07 | 中国科学技术大学 | High-performance human face recognition method and system |
| CN104504410A (en) * | 2015-01-07 | 2015-04-08 | 深圳市唯特视科技有限公司 | Three-dimensional face recognition device and method based on three-dimensional point cloud |
| CN105654035A (en) * | 2015-12-21 | 2016-06-08 | 湖南拓视觉信息技术有限公司 | Three-dimensional face recognition method and data processing device applying three-dimensional face recognition method |
Non-Patent Citations (4)
| Title |
|---|
| NAOUFEL WERGHI ET AL.: "The Mesh-LBP: A Framework for Extracting Local Binary Patterns From Discrete Manifolds", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| 吕士文: "基于改进BLP特征的三维人脸识别方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 郭鹏飞 等: "三维人脸识别系统在监狱出入口管理系统中的应用", 《中国安防》 * |
| 魏尚: "基于Kinect深度图像的三维人脸识别技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110069968A (en) * | 2018-01-22 | 2019-07-30 | 耐能有限公司 | Face recognition and face recognition method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104866829B (en) | A kind of across age face verification method based on feature learning | |
| CN110348319B (en) | Face anti-counterfeiting method based on face depth information and edge image fusion | |
| CN103942577B (en) | Based on the personal identification method for establishing sample database and composite character certainly in video monitoring | |
| CN106295124B (en) | The method of a variety of image detecting technique comprehensive analysis gene subgraph likelihood probability amounts | |
| CN109101938B (en) | Multi-label age estimation method based on convolutional neural network | |
| CN101540000B (en) | Iris classification method based on texture primitive statistical characteristic analysis | |
| CN109344825A (en) | A kind of licence plate recognition method based on convolutional neural networks | |
| CN109635875A (en) | A kind of end-to-end network interface detection method based on deep learning | |
| CN105825183B (en) | Facial expression recognizing method based on partial occlusion image | |
| CN108921201B (en) | Dam defect identification and classification method based on feature combination with CNN | |
| CN109190665A (en) | A kind of general image classification method and device based on semi-supervised generation confrontation network | |
| CN109902590A (en) | Pedestrian's recognition methods again of depth multiple view characteristic distance study | |
| CN107480575A (en) | The training method of model, across age face identification method and corresponding device | |
| CN104143079A (en) | Method and system for face attribute recognition | |
| CN103971106B (en) | Various visual angles facial image gender identification method and device | |
| CN103258204A (en) | Automatic micro-expression recognition method based on Gabor features and edge orientation histogram (EOH) features | |
| CN102332086A (en) | Facial identification method based on dual threshold local binary pattern | |
| CN105138968A (en) | Face authentication method and device | |
| CN103473571A (en) | Human detection method | |
| CN112052772A (en) | Face shielding detection algorithm | |
| CN101739555A (en) | Method and system for detecting false face, and method and system for training false face model | |
| CN107220598B (en) | Iris image classification method based on deep learning features and Fisher Vector coding model | |
| CN106778489A (en) | The method for building up and equipment of face 3D characteristic identity information banks | |
| CN106599785A (en) | Method and device for building human body 3D feature identity information database | |
| CN107239777A (en) | A kind of tableware detection and recognition methods based on various visual angles graph model |
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: 20171010 |