CN104156700A - Face image glass removal method based on mobile shape model and weighted interpolation method - Google Patents
Face image glass removal method based on mobile shape model and weighted interpolation method Download PDFInfo
- Publication number
- CN104156700A CN104156700A CN201410366526.1A CN201410366526A CN104156700A CN 104156700 A CN104156700 A CN 104156700A CN 201410366526 A CN201410366526 A CN 201410366526A CN 104156700 A CN104156700 A CN 104156700A
- Authority
- CN
- China
- Prior art keywords
- glasses
- region
- algorithm
- shape model
- face
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a face image glass removal method based on a mobile shape model and a weighted interpolation method, and the application of the face image glass removal method in face identification. According to the method, glass removal processing is performed on face data detected through the face detection technology, the processed data is used for face identification, and the accurate glass removal method can effectively improve the face identification accuracy. The mobile shape model is utilized by an algorithm for positioning a glass area, and the glass area is removed through the weighted-interpolation-based method, so that the glass removal effect is ensured. With adoption of the method, the problem that the recognition rate is reduced greatly because the deep color thick-frame glasses shield faces during the face identification process is effectively solved, therefore, the face identification performance is improved.
Description
Technical field
The present invention relates to computer vision field, relate in particular to a kind of method that facial image glasses are removed.
Background technology
Recognition of face is as an important research field in recent years, although obtained remarkable progress, but in the application of some reality, the many factors such as illumination, attitude, glasses exert an influence to recognition effect to some extent, and wherein glasses are more common a kind of chaff interferences.
Present stage, the most frequently used facial image glasses removal method was principal component analysis (PCA).The method is utilized glasses-free facial image training characteristics space, better for the input picture effect close with training image, but even cause discrimination to decline for easily introducing much noise with the quite different input picture of training image, and need the picture training of certain hour and some.
Summary of the invention
The invention provides a kind of new facial image glasses removal method, improved efficiency and effect that glasses are removed.
The present invention adopts following technical scheme:
Facial image glasses removal method based on moving shape model and weighted interpolation method, comprising:
(1) human face characteristic point that utilizes moving shape model to produce is oriented the initial search area of glasses;
(2) utilize the textural characteristics based on skin gray level co-occurrence matrixes in initial search area, to extract binary search region;
(3) in binary search region, obtain each connected domain area, think glasses region if be more than or equal to certain threshold value, think and be not glasses region if be less than this threshold value;
(4) the weighted interpolation method of carrying out in glasses region based on distance is removed glasses.
Compared with the conventional method, facial image glasses removal method disclosed in this invention, can save the training time, the input picture of single sample is carried out in real time to effectively glasses and remove.
Brief description of the drawings
Fig. 1 is the process flow diagram of moving shape model location glasses initial search area;
Fig. 2 is the process flow diagram in texture feature extraction glasses binary search region;
Fig. 3 is general flow chart.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
Embodiment is:
(1), to input picture gray processing, utilize the Adaboost detection of classifier face based on Haar feature;
(2) by reference to the accompanying drawings 1, utilize moving shape model to extract 68 positioning feature point in face and go out the region between eyebrow and face, then in this region, get rid of the region of nose, thus the initial search area of location glasses;
(3) by reference to the accompanying drawings 2, ask for based on gray level co-occurrence matrixes face dermatoglyph feature, and by finding in initial search area with face dermatoglyph characteristic matching degree lower than 80% position as binary search region;
(4) pinpoint target extraction is carried out to by the connected domain method based on gray-scale value in binary search region, in the time that pinpoint target area is more than or equal to 50 pixel, this target is defined as to glasses region; Otherwise, be defined as non-glasses region;
(5) in glasses region, carry out the weighted interpolation method based on distance that size is 5*5 and remove glasses, weight coefficient is Gaussian distribution and increases and reduce with distance.
Claims (6)
1. the facial image glasses removal method based on moving shape model and weighted interpolation method, is characterized in that utilizing moving shape model to carry out the location in glasses region, to improve efficiency of algorithm; Utilize and carry out the removal in glasses region based on the method for weighted interpolation, with the effect that ensures that glasses are removed.
2. method according to claim 1, it is characterized in that orienting the region between eyebrow and face based on the human face characteristic point of moving shape model generation, be the general regions that exist of glasses, then in this region, get rid of the region of nose, be generally domain of the existences not of glasses, thereby locate fast the initial search area of glasses.
3. method according to claim 1, it is characterized in that carrying out based on moving shape model the location in glasses region, its algorithm is, draw by algorithm claimed in claim 2 after the initial search area of glasses, by finding with the unmatched position of face dermatoglyph feature as binary search region in this region, then pinpoint target extraction is carried out to by the connected domain method based on gray-scale value in this region, in the time that pinpoint target size is in certain zone of reasonableness, this target is defined as to glasses region; If target sizes is directly filtered beyond the setting range time, experimental results show that this localization method can reduce algorithm computational complexity, thereby effectively improve efficiency of algorithm.
4. method according to claim 3, it is characterized in that the feature for texture in second extraction, algorithm obtains gray level co-occurrence matrixes by the glasses region to first extraction and near people's face skin corresponding region simultaneously to carry out, then by contrasting the entropy in gray level co-occurrence matrixes, energy weighted value confirms in the preliminary human face region extracting that whether certain position is as further screening region.
5. method according to claim 1, it is characterized in that carrying out glasses zone location algorithm based on moving shape model, its algorithm carries out the removal in glasses region by the method based on field interpolation, to ensure the effect of algorithm, show in algorithm and carry out pixel replacement by the pixel weighted interpolation method of glasses region being carried out based on distance, thereby ensure the effect that glasses are removed.
6. method according to claim 5, it is characterized in that after glasses extracted region, algorithm is by copying to the mean value of human face region pixel value in the certain limit of glasses region in this region, then utilize the pixel weighted interpolation method based on distance to adjust the pixel value in glasses region, thereby make the rear former glasses of glasses removal region approach the effect of people's face skin.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410366526.1A CN104156700A (en) | 2014-07-26 | 2014-07-26 | Face image glass removal method based on mobile shape model and weighted interpolation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410366526.1A CN104156700A (en) | 2014-07-26 | 2014-07-26 | Face image glass removal method based on mobile shape model and weighted interpolation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104156700A true CN104156700A (en) | 2014-11-19 |
Family
ID=51882197
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410366526.1A Pending CN104156700A (en) | 2014-07-26 | 2014-07-26 | Face image glass removal method based on mobile shape model and weighted interpolation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104156700A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105046250A (en) * | 2015-09-06 | 2015-11-11 | 广州广电运通金融电子股份有限公司 | Glasses elimination method for face recognition |
CN106846348A (en) * | 2017-02-16 | 2017-06-13 | 河北大学 | The method that glasses are automatically removed in facial image |
CN107424204A (en) * | 2017-06-28 | 2017-12-01 | 浙江工商大学 | Isomorphism Triangulation Algorithm based on gradual Planar Mapping |
CN107506708A (en) * | 2017-08-14 | 2017-12-22 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
CN107644228A (en) * | 2017-09-21 | 2018-01-30 | 联想(北京)有限公司 | Image processing method |
WO2018041237A1 (en) * | 2016-08-31 | 2018-03-08 | 腾讯科技(深圳)有限公司 | Human face authentication method and device, and storage medium |
CN107844742A (en) * | 2017-09-26 | 2018-03-27 | 平安科技(深圳)有限公司 | Facial image glasses minimizing technology, device and storage medium |
WO2018072102A1 (en) * | 2016-10-18 | 2018-04-26 | 华为技术有限公司 | Method and apparatus for removing spectacles in human face image |
US11830209B2 (en) | 2017-05-26 | 2023-11-28 | Snap Inc. | Neural network-based image stream modification |
US11887260B2 (en) | 2021-12-30 | 2024-01-30 | Snap Inc. | AR position indicator |
US11928783B2 (en) | 2021-12-30 | 2024-03-12 | Snap Inc. | AR position and orientation along a plane |
US11954762B2 (en) | 2022-01-19 | 2024-04-09 | Snap Inc. | Object replacement system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1949246A (en) * | 2006-11-08 | 2007-04-18 | 中山大学 | Multiple expression whole face profile testing method based on moving shape model |
KR20090093223A (en) * | 2008-02-29 | 2009-09-02 | 홍익대학교 산학협력단 | Removal Eye Glasses using Variable Mask and Inpainting for Improved Performance of Face Recognition System |
CN102034079A (en) * | 2009-09-24 | 2011-04-27 | 汉王科技股份有限公司 | Method and system for identifying faces shaded by eyeglasses |
CN103020579A (en) * | 2011-09-22 | 2013-04-03 | 上海银晨智能识别科技有限公司 | Face recognition method and system, and removing method and device for glasses frame in face image |
CN103839223A (en) * | 2012-11-21 | 2014-06-04 | 华为技术有限公司 | Image processing method and image processing device |
-
2014
- 2014-07-26 CN CN201410366526.1A patent/CN104156700A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1949246A (en) * | 2006-11-08 | 2007-04-18 | 中山大学 | Multiple expression whole face profile testing method based on moving shape model |
KR20090093223A (en) * | 2008-02-29 | 2009-09-02 | 홍익대학교 산학협력단 | Removal Eye Glasses using Variable Mask and Inpainting for Improved Performance of Face Recognition System |
CN102034079A (en) * | 2009-09-24 | 2011-04-27 | 汉王科技股份有限公司 | Method and system for identifying faces shaded by eyeglasses |
CN103020579A (en) * | 2011-09-22 | 2013-04-03 | 上海银晨智能识别科技有限公司 | Face recognition method and system, and removing method and device for glasses frame in face image |
CN103839223A (en) * | 2012-11-21 | 2014-06-04 | 华为技术有限公司 | Image processing method and image processing device |
Non-Patent Citations (3)
Title |
---|
XIAODONG JIA ETC: "Eyeglasses Removal From Facial Image Based On Phase Congruency", 《2013 3RD INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING》 * |
杨秋芬等: "驾驶员疲劳驾驶中的眼睛定位创新算法", 《计算机工程与应用》 * |
郭航等: "灰度共生矩阵在皮肤纹理检测中的应用研究", 《中国图象图形学报》 * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017036160A1 (en) * | 2015-09-06 | 2017-03-09 | 广州广电运通金融电子股份有限公司 | Glasses removal method for facial recognition |
CN105046250A (en) * | 2015-09-06 | 2015-11-11 | 广州广电运通金融电子股份有限公司 | Glasses elimination method for face recognition |
WO2018041237A1 (en) * | 2016-08-31 | 2018-03-08 | 腾讯科技(深圳)有限公司 | Human face authentication method and device, and storage medium |
US10922529B2 (en) | 2016-08-31 | 2021-02-16 | Tencent Technology (Shenzhen) Company Limited | Human face authentication method and apparatus, and storage medium |
CN109416727A (en) * | 2016-10-18 | 2019-03-01 | 华为技术有限公司 | Glasses minimizing technology and device in a kind of facial image |
WO2018072102A1 (en) * | 2016-10-18 | 2018-04-26 | 华为技术有限公司 | Method and apparatus for removing spectacles in human face image |
CN106846348B (en) * | 2017-02-16 | 2019-07-12 | 河北大学 | The method of glasses is automatically removed in facial image |
CN106846348A (en) * | 2017-02-16 | 2017-06-13 | 河北大学 | The method that glasses are automatically removed in facial image |
US11830209B2 (en) | 2017-05-26 | 2023-11-28 | Snap Inc. | Neural network-based image stream modification |
CN107424204B (en) * | 2017-06-28 | 2020-09-01 | 浙江工商大学 | Isomorphic triangulation method based on progressive plane mapping |
CN107424204A (en) * | 2017-06-28 | 2017-12-01 | 浙江工商大学 | Isomorphism Triangulation Algorithm based on gradual Planar Mapping |
CN107506708A (en) * | 2017-08-14 | 2017-12-22 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
CN107644228B (en) * | 2017-09-21 | 2020-05-26 | 联想(北京)有限公司 | Image processing method |
CN107644228A (en) * | 2017-09-21 | 2018-01-30 | 联想(北京)有限公司 | Image processing method |
WO2019061659A1 (en) * | 2017-09-26 | 2019-04-04 | 平安科技(深圳)有限公司 | Method and device for removing eyeglasses from facial image, and storage medium |
CN107844742B (en) * | 2017-09-26 | 2019-01-04 | 平安科技(深圳)有限公司 | Facial image glasses minimizing technology, device and storage medium |
CN107844742A (en) * | 2017-09-26 | 2018-03-27 | 平安科技(深圳)有限公司 | Facial image glasses minimizing technology, device and storage medium |
US11887260B2 (en) | 2021-12-30 | 2024-01-30 | Snap Inc. | AR position indicator |
US11928783B2 (en) | 2021-12-30 | 2024-03-12 | Snap Inc. | AR position and orientation along a plane |
US11954762B2 (en) | 2022-01-19 | 2024-04-09 | Snap Inc. | Object replacement system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104156700A (en) | Face image glass removal method based on mobile shape model and weighted interpolation method | |
CN103310218B (en) | A kind of overlap blocks fruit precise recognition method | |
Yu et al. | Vision-based hand gesture recognition using combinational features | |
Thalji et al. | Iris Recognition using robust algorithm for eyelid, eyelash and shadow avoiding | |
EP3111421A1 (en) | Eye gaze tracking | |
CN103870808A (en) | Finger vein identification method | |
CN103810466A (en) | Apparatus and method for face recognition | |
CN103218605A (en) | Quick eye locating method based on integral projection and edge detection | |
CN103336946A (en) | Binocular stereoscopic vision based clustered tomato identification method | |
CN104573634A (en) | Three-dimensional face recognition method | |
Tabassum et al. | Text detection using MSER and stroke width transform | |
CN108710883B (en) | Complete salient object detection method adopting contour detection | |
Huang et al. | Text detection and recognition in natural scene images | |
CN111191569A (en) | Face attribute recognition method and related device thereof | |
CN111353385A (en) | Pedestrian re-identification method and device based on mask alignment and attention mechanism | |
WO2017101380A1 (en) | Method, system, and device for hand recognition | |
Lee et al. | A Hand gesture recognition system based on difference image entropy | |
CN116434071B (en) | Determination method, determination device, equipment and medium for normalized building mask | |
Das et al. | Human face detection in color images using HSV color histogram and WLD | |
CN104156720A (en) | Face image denoising method on basis of noise evaluation model | |
CN104156719A (en) | Face image light processing method on basis of shape and light model | |
Liu et al. | Identification Method of Sunflower Leaf Disease Based on SIFT Point | |
Kusuma et al. | Skeletonization using thinning method for human motion system | |
Wang et al. | A new method of speeded up robust features image registration based on image preprocessing | |
CN109145875B (en) | Method and device for removing black frame glasses in face image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141119 |
|
WD01 | Invention patent application deemed withdrawn after publication |