CN100375977C - Three-dimensional portrait imaging device and distinguishing method for three-dimensional human face - Google Patents
Three-dimensional portrait imaging device and distinguishing method for three-dimensional human face Download PDFInfo
- Publication number
- CN100375977C CN100375977C CNB2005100248281A CN200510024828A CN100375977C CN 100375977 C CN100375977 C CN 100375977C CN B2005100248281 A CNB2005100248281 A CN B2005100248281A CN 200510024828 A CN200510024828 A CN 200510024828A CN 100375977 C CN100375977 C CN 100375977C
- Authority
- CN
- China
- Prior art keywords
- image
- dimensional
- face
- human face
- discrimination
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a three-dimensional human face discrimination method which is used for judging whether a human face image obtained by a video device is a three-dimensional image or not. Firstly, images of the left side and the right side of the human face are simultaneously obtained by the video device, then, FL and FR of the images of the left side and the right side of the human face are cut out, subsequently, the images of the human face at both sides are processed in a normalization way to form an image with a certain resolution in a masking way, furthermore, the FL and the FR images are correspondingly cut into a plurality of sheets, in addition, mean square deviations D1, D2 and D3 of the pixel difference of the cut sheets in corresponding positions of the FL and the FR are calculated, and finally, according to the values of the mean square deviations, the method judges whether the image is a three-dimensional image or not. The method can conveniently judge whether a human face image is a three-dimensional image or not and can prevent the human face from being impersonated by pictures or graphics in some occasions.
Description
Technical field
The present invention relates to the living things feature recognition field, the especially relevant a kind of people's of judgement face is the method for discrimination of the three-dimensional face or the three-dimensional face of planar image.
Background technology
Face recognition technology is as a hot issue of field of biological recognition, and a lot of research institutions and company be all in positive studying and corresponding product development, and worked out multiple face recognition algorithms.But product all obtains portrait based on 2 dimensional imaging equipment at present, the depth information of nobody's picture, can not distinguish the human face portrait and real three-dimensional face picture of two dimension, will can not judge exactly whether the image that is in monitoring or the miscellaneous equipment is real people's face, has limited The Application of Technology in some occasion like this.
Next video camera of normal conditions can only shape subject image under at an angle, we can't know 3 dimension information of object.Fields such as modern spacer remote sensing, medical imaging adopt scanning techniques such as laser, ray to realize 3 dimension information acquisitions, and are with high costs, can't enter large-scale commercial the application.
Summary of the invention
The objective of the invention is at the deficiencies in the prior art, a kind of method of discrimination of three-dimensional face that can distinguish real three-dimensional face and the image of two dimension easily is provided.
The invention provides a kind of three-dimensional face method of discrimination, in order to judge whether the people's face that obtains by camera head is 3-D view, may further comprise the steps:
The first step is obtained the image of the left and right sides of people's face simultaneously by camera head;
Second step, determine the eye position of the right and left imaging by eye position location algorithm, cut out the image FL and the FR of corresponding people from left and right sides face;
In the 3rd step, the facial image of both sides is carried out normalized and mask forms the image with certain resolution;
The 4th step was cut into some with described FL and FR image correspondence, obtained the mean square deviation of the pixel value difference of the correspondence position on the respective slice among FL and the FR, judged according to the numerical value of described mean square deviation whether image is the image that three-dimensional face forms then.
Characteristics of the present invention also are: described camera head is a three-dimensional portrait imaging device, comprise a camera, two active reflective mirrors and two passive reflective mirrors, the minute surface of wherein said active reflective mirror is towards the people's face that needs capture, opposing and the setting that forms an angle of described two passive reflective mirrors, and it is parallel with the active reflective mirror, minute surface is relative, and the capture scope of described camera is the minute surface of two passive reflective mirrors.
Compared with prior art, the invention has the beneficial effects as follows: can distinguish real three-dimensional face and two dimensional image easily.
Description of drawings
Fig. 1 is the structural representation of an embodiment of three-dimensional portrait imaging device of the present invention.
Fig. 2 is the light-dividing principle figure of an embodiment of three-dimensional portrait imaging device of the present invention.
Fig. 3 is to use the imaging effect figure of three-dimensional portrait imaging device imaging of the present invention.
Fig. 4 is the process flow diagram of three-dimensional face method of discrimination of the present invention.
Fig. 5 (a) and Fig. 5 (b) are the design sketchs of three-dimensional face method of discrimination of the present invention, in order to show from a formed real human face image of Fig. 3 and a two-dimension picture contrast effect behind the detection and location mask.
Embodiment
Below in conjunction with drawings and Examples the present invention is further described.
See also shown in Figure 1, the structural representation of a kind of three-dimensional portrait imaging device of the present invention, three-dimensional portrait imaging device comprises a camera 1 and two groups initiatively reflective mirror 2 and passive reflective mirror 3, initiatively the minute surface of reflective mirror 2 is towards the portrait that needs capture, described passive reflective mirror 3 is opposing and become the setting of certain right angle, and parallel with active reflective mirror 2, minute surface is relative, and the capture scope of described camera 1 is the minute surface of a plurality of passive reflective mirrors 3.
See also shown in Figure 2, it is the light-dividing principle figure of a kind of three-dimensional portrait imaging device of the present invention, because two groups of initiatively reflective mirror 2 and passive reflective mirror 3 left-right symmetric, a pair of active reflective mirror 2 by the right and left reflexes to middle a pair of passive reflective mirror 3 with image in kind respectively, and passive then reflective mirror 3 reflexes to camera 1 with the image in kind of the left and right sides.Like this, see image in the past with regard to the both sides that formed same object in the camera 1, its image effect as shown in Figure 3.
See also shown in Fig. 4 and Fig. 5 (a) and Fig. 5 (b), the process flow diagram of a kind of three-dimensional face method of discrimination of the present invention and imaging effect figure, in order to judge whether the facial image that obtains by camera head is 3-D view, may further comprise the steps:
The first step is the image that three-dimensional portrait imaging device obtains the left and right sides of people's face simultaneously by camera head;
Second step, determine the eye position of the right and left imaging by eye position location algorithm, cut out the image FL and the FR of corresponding people from left and right sides face;
In the 3rd step, the facial image of both sides is carried out normalized and mask, and to form resolution be the image of 64*64;
In the 4th step, FL, FR image transversally cutting are become three (width is respectively 18,28,18);
The 5th goes on foot, and obtains mean square deviation D1, D2, the D3 of the pixel value difference of the section of correspondence position among FL and the FR, promptly deducts the pixel value of first the corresponding point of FR with each pixel value of first of FL, obtains mean square deviation D1; D2 and D3 analogize; The mean square deviation D1=19.8 that the image of use real human face is tried to achieve; D2=26.7; D3=25.3, and use two-dimensional image D1=8.0; D2=10.8; D3=12.2.
The numerical value of the 6th step according to mean square deviation judges whether to be 3-D view, when 15＜D1＜30 and 15＜D2＜30 and 15＜D3＜30, be judged as three-dimensional, when 5＜D1＜15 or 5＜D2＜15 or 5＜D3＜15, be judged as two dimension, other situation then is an error image.Thus, process decision chart 5 (a) is depicted as three-dimensional face, and Fig. 5 (b) is depicted as two dimensional image.
In sum, the present invention has finished inventor's goal of the invention, can realize the differentiation of three-dimensional face easily, uses picture personation people face, and only uses a camera to finish, and is with low cost, need not the high scanning device of other price.
Claims (5)
1. three-dimensional face method of discrimination, in order to judge whether the people's face that obtains by camera head is 3-D view, may further comprise the steps:
The first step is obtained the image of the left and right sides of people's face simultaneously by camera head;
Second step, determine the eye position of the right and left imaging by eye position location algorithm, cut out the image FL and the FR of corresponding people from left and right sides face;
In the 3rd step, the facial image of both sides is carried out normalized and mask forms the image with certain resolution;
The 4th step was cut into some with described FL and FR image correspondence, obtained the mean square deviation of the pixel value difference of the correspondence position on the respective slice among FL and the FR, judged according to the numerical value of described mean square deviation whether image is the image that three-dimensional face forms then.
2. the method for discrimination of three-dimensional face as claimed in claim 1 is characterized in that: the cutting in described the 4th step is cut into three.
3. the method for discrimination of three-dimensional face as claimed in claim 2, it is characterized in that: if D1, D2, D3 be in 15 and 30 between differentiate and be 3-D view, the mean square deviation of each respective slice pixel value difference when wherein D1, D2, D3 are FL and the cutting of FR image be cut into three.
4. the method for discrimination of three-dimensional face as claimed in claim 1 is characterized in that: described the 3rd step intermediate-resolution is 64*64.
5. the method for discrimination of three-dimensional face as claimed in claim 1, it is characterized in that: described camera head is a three-dimensional portrait imaging device, comprise a camera, two active reflective mirrors and two passive reflective mirrors, the minute surface of wherein said active reflective mirror is towards the people's face that needs capture, opposing and the setting that forms an angle of described two passive reflective mirrors, and parallel with the active reflective mirror, minute surface is relative, and the capture scope of described camera is the minute surface of two passive reflective mirrors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100248281A CN100375977C (en) | 2005-04-01 | 2005-04-01 | Three-dimensional portrait imaging device and distinguishing method for three-dimensional human face |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100248281A CN100375977C (en) | 2005-04-01 | 2005-04-01 | Three-dimensional portrait imaging device and distinguishing method for three-dimensional human face |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1841405A CN1841405A (en) | 2006-10-04 |
| CN100375977C true CN100375977C (en) | 2008-03-19 |
Family
ID=37030417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100248281A Expired - Fee Related CN100375977C (en) | 2005-04-01 | 2005-04-01 | Three-dimensional portrait imaging device and distinguishing method for three-dimensional human face |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100375977C (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393597B (en) * | 2007-09-19 | 2011-06-15 | 上海银晨智能识别科技有限公司 | Method for identifying front of human face |
| CN101609500B (en) * | 2008-12-01 | 2012-07-25 | 公安部第一研究所 | Quality estimation method of exit-entry digital portrait photos |
| CN102103696A (en) * | 2009-12-21 | 2011-06-22 | 鸿富锦精密工业（深圳）有限公司 | Face identification system, method and identification device with system |
| CN104573682A (en) * | 2015-02-15 | 2015-04-29 | 四川川大智胜软件股份有限公司 | Face anti-counterfeiting method based on face similarity |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5828913A (en) * | 1995-06-06 | 1998-10-27 | Zanen; Pieter O. | Method for three dimensional measurement and imaging having focus-related convergence compensation |
| WO2004072899A1 (en) * | 2003-02-13 | 2004-08-26 | Nec Corporation | Unauthorized person detection device and unauthorized person detection method |
-
2005
- 2005-04-01 CN CNB2005100248281A patent/CN100375977C/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5828913A (en) * | 1995-06-06 | 1998-10-27 | Zanen; Pieter O. | Method for three dimensional measurement and imaging having focus-related convergence compensation |
| WO2004072899A1 (en) * | 2003-02-13 | 2004-08-26 | Nec Corporation | Unauthorized person detection device and unauthorized person detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1841405A (en) | 2006-10-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109598242B (en) | Living body detection method | |
| CN105574525B (en) | A kind of complex scene multi-modal biological characteristic image acquiring method and its device | |
| KR101355974B1 (en) | Method and devices for tracking multiple object | |
| CN109643366A (en) | For monitoring the method and system of the situation of vehicle driver | |
| EP2597597B1 (en) | Apparatus and method for calculating three dimensional (3D) positions of feature points | |
| CN106485735A (en) | Human body target recognition and tracking method based on stereovision technique | |
| CN104021538B (en) | Object positioning method and device | |
| CN102930334B (en) | Video recognition counter for body silhouette | |
| JP5127531B2 (en) | Image monitoring device | |
| EP2717013A1 (en) | Stereo camera device and computer-readable recording medium | |
| CN105740775A (en) | Three-dimensional face living body recognition method and device | |
| CN104915656B (en) | A kind of fast human face recognition based on Binocular vision photogrammetry technology | |
| CN107909604A (en) | Dynamic object movement locus recognition methods based on binocular vision | |
| Hongo et al. | Focus of attention for face and hand gesture recognition using multiple cameras | |
| JP6590609B2 (en) | Image analysis apparatus and image analysis method | |
| CN105913013A (en) | Binocular vision face recognition algorithm | |
| CN105740778A (en) | Improved three-dimensional human face in-vivo detection method and device thereof | |
| Hossny et al. | Low cost multimodal facial recognition via kinect sensors | |
| CN100375977C (en) | Three-dimensional portrait imaging device and distinguishing method for three-dimensional human face | |
| CN101876535A (en) | Method, device and monitoring system for height measurement | |
| CN105611142A (en) | Shooting method and apparatus thereof | |
| CN112926464A (en) | Face living body detection method and device | |
| CN112633217A (en) | Human face recognition living body detection method for calculating sight direction based on three-dimensional eyeball model | |
| CN112801038B (en) | Multi-view face in-vivo detection method and system | |
| CN105931429A (en) | Intelligent nighttime approach recognition and alarming method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080319 Termination date: 20140401 |