CN103942525A - Real-time face optimal selection method based on video sequence - Google Patents
Real-time face optimal selection method based on video sequence Download PDFInfo
- Publication number
- CN103942525A CN103942525A CN201310737499.XA CN201310737499A CN103942525A CN 103942525 A CN103942525 A CN 103942525A CN 201310737499 A CN201310737499 A CN 201310737499A CN 103942525 A CN103942525 A CN 103942525A
- Authority
- CN
- China
- Prior art keywords
- face
- maxscore
- image
- value
- 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
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a real-time face optimal selection method based on a video sequence. The method comprises the following steps: a face image is acquired; face clarity, face size and opening degree of human eyes act as three indexes for face quality evaluation so that a face integrated evaluation score ImageScore is calculated; the ImageScore of the first frame of the face image is assigned to an initial face optimal image MaxScore and the face image is stored; the MaxScore after assignment is compared with the ImageScore of other frames of the face image, and the face image corresponding to the higher ImageScore is stored and the ImageScore is assigned to the MaxScore again; and the MaxScore value is cyclically updated and the stored optimal face image is updated until the image sequence with the face cannot be acquired, and the cycle is ended and the image corresponding to the stored MaxScore is the optimal face image. The face image with higher quality can be screened for identification via comparison of the integrated evaluation indexes of face quality in the video sequence under the situation of not knowing the face sequence image in advance; besides, the calculation method is simple with real-time performance.
Description
Technical field
The present invention relates to a kind of image quality evaluating method, particularly relate to a kind of real-time face method for optimizing based on video sequence.
Background technology
Along with the development of computer vision technique, face recognition technology has been widely used in a lot of electronic systems, as gate control system, Gate System, E-Passport, public security, bank self-help system, information security etc.But how the situation that easily produces mistake identification in face recognition process, improve discrimination and just seem particularly important.Identify with the facial image that a width is of high quality, can produce higher matching rate, thereby improve recognition of face rate.
In order to filter out the good facial image of quality from sequence image, just need to evaluate the quality of facial image, pick out the highest facial image of scoring.Current image quality evaluating method is mainly divided into subjective evaluation method and method for objectively evaluating.Subjective evaluation method is by observer's subjective feeling, image to be marked, thereby judges the quality of picture quality.Method for objectively evaluating is the subjective vision system made mathematical model according to human eye, and by concrete formula computed image quality.But because subjective evaluation method cost is high, the cycle is long, so current image quality evaluating method mainly adopts method for objectively evaluating.
Also there is no at present Patents about facial image method for optimizing, only have minority paper to relate to, as be entitled as the paper of " quality of human face image evaluation assessment ".The method that this paper relates to is a kind of method of evaluating based on many indexs, belongs to the method for objectively evaluating of picture quality.By the coefficient such as brightness and the sharpness index comprehensive evaluation facial image of face size, face angle, picture contrast, image, but need to calculate multiple indexs, computing method complexity also has the requirement of index weights simultaneously.
Summary of the invention
In order to solve, subjective evaluation method cost is high, the cycle is long, and current method for objectively evaluating exists the problem of computing method complexity, and the present invention adopts method for objectively evaluating, and a kind of real-time face method for optimizing based on video sequence is provided.
In order to achieve the above object, the technical solution adopted in the present invention is as follows:
A real-time face method for optimizing based on video sequence, comprises following steps:
1) gather facial image, face is gone out to a series of images gathering between now to face disappearance moment as an image sequence, from sequence image, obtain facial image;
2) first index using face sharpness as face quality assessment, adopts energy gradient function to calculate face sharpness G;
3) second index using the size of face as face quality assessment, by calculating the area evaluator of the human face region little FaceSize that is bold;
4) the 3rd index using human eye opening degree as face quality assessment, behind the candidate region of selected human eye, adopt the method for vertical projection to carry out bridge of the nose location, thereby the candidate region of human eye is divided into left and right two parts, then adopt otsu threshold method to carry out binaryzation, cut apart and obtain human eye area, and then add up the number of pixels that in this human eye area, pixel value is 0, the measurement index using this statistical value EyeNum as human eye opening degree;
5) according to step 2) to G described in step 4), FaceSize, EyeNum calculating face comprehensive evaluation mark ImageScore.The standard good due to quality of human face image is: sharpness is high, and face is large, eyes opening degree is high.Therefore this programme proposes to adopt above three kinds of indexs facial image to be carried out to comprehensive evaluation, i.e. sharpness, face size and human eye opening degree.Above three kinds of indexs suitable for the preferred importance of face, in order to reduce the complexity of determining weighted value, therefore comprehensive evaluation index is obtained by the arithmetic sum of above three kinds of indexs;
6) the comprehensive evaluation mark MaxScore initial value of setting best face is 0, utilize above-mentioned ImageScore to give a mark to the first frame facial image, assignment is in MaxScore and preserve facial image, then utilize above-mentioned ImageScore to give a mark to the second frame facial image collecting, compare with MaxScore value, be greater than MaxScore value by this ImageScore assignment to MaxScore, and preserve its facial image; Otherwise MaxScore value and the facial image of preserving are constant;
7) utilize above-mentioned ImageScore to give a mark to this facial image of other frames in the sequence image collecting successively, and compare with MaxScore value respectively, if be greater than MaxScore value, this ImageScore assignment, to MaxScore, and is preserved to its facial image; Otherwise MaxScore value and the facial image of preserving are constant;
8) circulation upgrade MaxScore value and upgrade preserve best facial image, until do not collect existence this face image sequence time, end loop, the best facial image that the image corresponding to MaxScore value of preservation is this face.
Further, the human face region more than lip due to human eye and the bridge of the nose, so in step 4) first using upper 2/3 part of human face region as human eye candidate region.Because human eye is about bridge of the nose symmetry, nasal bridge region is two the brightest parts in centre, so can determine by the method for vertical projection the position of the bridge of the nose, human eye candidate region is divided into left and right two parts.
Further, carry out human eye area pixels statistics, specifically refer to taking bridge of the nose position as with reference to the value in the right and left eyes region after binaryzation is added up to summation as 0 number of pixels.By the folding degree of EyeNum assessment human eye, EyeNum is larger, and to show that human eye is opened larger; Otherwise EyeNum is less, and to show that human eye is opened less, in eyes closed situation, number of pixels is relatively minimum.
Preferably, because human eye area is the region that gray-scale value is relatively little in whole human face region, so this programme adopts otsu threshold method to ask binary-state threshold, the image of human face region is carried out to binaryzation.Maximum variance between clusters (being called for short otsu) is a kind of adaptive Threshold, and it is the gamma characteristic according to image, and image is divided into background and target two parts.Inter-class variance between background and target is larger, illustrates that this two parts difference is larger, and part target mistake is divided into background or part background mistake and is divided into target and all can causes both inter-class variances to diminish.Therefore, the error probability of the Threshold segmentation of inter-class variance maximum is less.Suppose that human eye area regards target as, other parts of human face region are regarded background as, because other part gray-scale values of human eye area and human face region differ larger, so can determine segmentation threshold by maximum between-cluster variance.
Further, described comprehensive evaluation mark ImageScore is obtained by the arithmetic sum of each index simply, has saved the time of calculating weighted value, has improved picture appraisal efficiency.Calculation equation is:
.The standard good due to quality of human face image is: sharpness is high, and face is large, eyes opening degree is high.Therefore this programme proposes to adopt above three kinds of indexs facial image to be carried out to comprehensive evaluation, i.e. sharpness, face size and human eye opening degree.Above three kinds of indexs suitable for the preferred importance of face, in order to reduce the complexity of determining weights, therefore comprehensive evaluation index directly carries out arithmetic sum by above three kinds of indexs and obtains.
The invention has the beneficial effects as follows: can, not knowing in advance in the sequence image situation of face, contrast and make evaluation by the face quality in video sequence, filter out facial image that quality is higher for identification.Three kinds of indexs that propose in scheme are all for facial image itself, and each index calculating method is simple, and overall target does not need training quota weight, can reach real-time.
Brief description of the drawings
Fig. 1 face preferred flow charts of the present invention;
Fig. 2 human eye opening degree of the present invention calculation flow chart;
Fig. 3 human eye opening degree result of calculation embodiment;
Fig. 4 adopts the inventive method to simulate the best face test result 1 obtaining;
Fig. 5 adopts the inventive method to simulate the best face test result 2 obtaining.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the invention is described further.As shown in Figure 1,
A real-time face method for optimizing based on video sequence, comprises following steps:
1) adopt USB camera collection video sequence, from sequence image, obtain facial image;
2) adopt energy gradient function to calculate face sharpness G, computing formula is
Wherein G represents the sharpness of human face region, presentation video gray-scale value, the scope of x, y is the scope of face rectangle frame;
3) calculate face size by the wide high product of human face region, computing formula is
FaceSize=w*h
Wherein, w and h represent respectively the wide and high of human face region;
4) as shown in Figure 2, assessment human eye opening degree, is divided into the following steps:
(a) upper 2/3 part that the candidate region of selected human eye is human face region, adopts the method for vertical projection to carry out bridge of the nose location, and formula is:
Its neutralization is respectively upper left corner coordinate and the lower right corner coordinate of candidate's human eye area, represents the gray-scale value at some place, and is respectively interior average integral vertical projection function and average variance vertical projection function.To above-mentioned two projection function normalization, and ask function MPV, formula is as follows:
Because two middle nasal bridge region are the brightest, so the maximum position of MPV value is the bridge of the nose, then according to bridge of the nose position, human eye candidate region is divided into left and right two parts;
(b) adopt otsu threshold method to ask binary-state threshold, binaryzation is carried out in human eye candidate region.Above-mentioned gained threshold value is as binary-state threshold, and the pixel value that is less than threshold value is 0, is judged as human eye area; The pixel value that is greater than threshold value is 255, is judged as non-human eye area.Statistical pixel values is 0 number of pixels EyeNum, the measurement index using this statistical value EyeNum as human eye opening degree.EyeNum is larger, and to show that human eye is opened larger, otherwise EyeNum is less, and to show that human eye is opened less, and in eyes closed situation, number of pixels is relatively minimum; Simulation human eye folding, comprise open eyes, micro-situation such as close, close one's eyes, gather two picture group sheets for human eye opening degree, method of the present invention to be tested, Fig. 3 is the test result obtaining according to step described in Fig. 2;
5) according to step 2) to G described in step 4), FaceSize, EyeNum calculating face comprehensive evaluation mark ImageScore, computing formula is:
；
6) the comprehensive evaluation mark MaxScore initial value of setting best face is 0, utilizes above-mentioned ImageScore to give a mark to the first frame facial image, and assignment is in MaxScore and preserve facial image.Then utilize above-mentioned ImageScore to give a mark to the second frame facial image collecting, with MaxScore value relatively, be greater than MaxScore value by this ImageScore assignment to MaxScore, and preserve its facial image; Otherwise MaxScore value and the facial image of preserving are constant;
7) utilize above-mentioned ImageScore to give a mark to this facial image of other frames in the sequence image collecting successively, and compare with MaxScore value respectively, if be greater than MaxScore value, this ImageScore assignment, to MaxScore, and is preserved to its facial image; Otherwise MaxScore value and the facial image of preserving are constant;
8) circulation upgrade MaxScore value and upgrade preserve best facial image, until do not collect existence this face image sequence time, end loop, the best facial image that the image corresponding to MaxScore value of preservation is this face.
specific embodiment:the situation of the motion from the close-by examples to those far off of simulation face, the motion that draws near, gather respectively three groups and comprise different face sequence images, adopt face method for optimizing of the present invention to test, Fig. 4 is the best face test result that from the close-by examples to those far off (50cm-150cm) obtains, the best face test result that Fig. 5 obtains for draw near (150cm-50cm).
The description of better case study on implementation is provided above, so that any technician of this area uses or utilize the present invention better, institute it should be understood that and the foregoing is only specific embodiments of the invention, is not limited to the present invention.To this better case study on implementation; those skilled in the art are not departing from the basis of the principle of the invention; can make various amendments or conversion, any amendment of making, be equal to replacement, improvement etc., be interpreted as these amendments or conversion does not depart from protection scope of the present invention.
Claims (4)
1. the real-time face method for optimizing based on video sequence, is characterized in that, the method comprises following steps:
1) gather facial image, face is gone out to a series of images gathering between now to face disappearance moment as an image sequence, from sequence image, obtain facial image;
2) first index using face sharpness as face quality assessment, adopts energy gradient function to calculate face sharpness G;
3) second index using the size of face as face quality assessment, by calculating the area evaluator of the human face region little FaceSize that is bold;
4) the 3rd index using human eye opening degree as face quality assessment, behind the candidate region of selected human eye, adopt the method for vertical projection to carry out bridge of the nose location, thereby the candidate region of human eye is divided into left and right two parts, then adopt otsu threshold method to carry out binaryzation, cut apart and obtain human eye area, and then add up the number of pixels that in this human eye area, pixel value is 0, the measurement index using this statistical value EyeNum as human eye opening degree;
5) according to step 2) to G described in step 4), FaceSize, EyeNum calculating face comprehensive evaluation mark ImageScore;
6) the comprehensive evaluation mark MaxScore initial value of setting best face is 0, utilize above-mentioned ImageScore to give a mark to the first frame facial image, assignment is in MaxScore and preserve facial image, then utilize above-mentioned ImageScore to give a mark to the second frame facial image collecting, compare with MaxScore value, be greater than MaxScore value by this ImageScore assignment to MaxScore, and preserve its facial image; Otherwise MaxScore value and the facial image of preserving are constant;
7) utilize above-mentioned ImageScore to give a mark to this facial image of other frames in the sequence image collecting successively, and compare with MaxScore value respectively, if be greater than MaxScore value, this ImageScore assignment, to MaxScore, and is preserved to its facial image; Otherwise MaxScore value and the facial image of preserving are constant;
8) circulation upgrade MaxScore value and upgrade preserve best facial image, until do not collect existence this face image sequence time, end loop, the best facial image that the image corresponding to MaxScore value of preservation is this face.
2. the real-time face method for optimizing based on video sequence according to claim 1, is characterized in that the human eye candidate region described in step 4) refers to upper 2/3 part of human face region.
3. the real-time face method for optimizing based on video sequence according to claim 1, it is characterized in that, described statistical value EyeNum specifically refers to taking bridge of the nose position as with reference to the value in right and left eyes region add up to summation as 0 number of pixels, and assesses the folding degree of human eye by the value that obtains EyeNum; EyeNum is larger, and to show that human eye is opened larger, otherwise EyeNum is less, and to show that human eye is opened less.
4. the real-time face method for optimizing based on video sequence according to claim 1, is characterized in that, described comprehensive evaluation mark ImageScore is by step 2) obtain to three kinds of index G, FaceSize in step 4), the arithmetic sum of EyeNum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310737499.XA CN103942525A (en) | 2013-12-27 | 2013-12-27 | Real-time face optimal selection method based on video sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310737499.XA CN103942525A (en) | 2013-12-27 | 2013-12-27 | Real-time face optimal selection method based on video sequence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103942525A true CN103942525A (en) | 2014-07-23 |
Family
ID=51190191
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310737499.XA Pending CN103942525A (en) | 2013-12-27 | 2013-12-27 | Real-time face optimal selection method based on video sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103942525A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554348A (en) * | 2015-12-25 | 2016-05-04 | 北京奇虎科技有限公司 | Image display method and device based on video information |
| WO2016164720A1 (en) * | 2015-04-10 | 2016-10-13 | Microsoft Technology Licensing, Llc | Snapshot capture for a communication session |
| CN106295585A (en) * | 2016-08-16 | 2017-01-04 | 深圳云天励飞技术有限公司 | A kind of filtration system of selection taking into account real-time and face quality and system |
| CN106446851A (en) * | 2016-09-30 | 2017-02-22 | 厦门大图智能科技有限公司 | Visible light based human face optimal selection method and system |
| CN107346426A (en) * | 2017-07-10 | 2017-11-14 | 深圳市海清视讯科技有限公司 | A kind of face information collection method based on video camera recognition of face |
| CN107358177A (en) * | 2017-06-27 | 2017-11-17 | 维拓智能科技（深圳）有限公司 | A kind of medium and long distance pedestrian detection method and terminal device based on graphical analysis |
| CN107633209A (en) * | 2017-08-17 | 2018-01-26 | 平安科技（深圳）有限公司 | Electronic installation, the method and storage medium of dynamic video recognition of face |
| CN107958220A (en) * | 2017-12-06 | 2018-04-24 | 杭州魔点科技有限公司 | A kind of face database compression processing method and its intelligent apparatus based on recognition of face |
| CN108388885A (en) * | 2018-03-16 | 2018-08-10 | 南京邮电大学 | The identification in real time of more people's features towards large-scale live scene and automatic screenshot method |
| CN108491784A (en) * | 2018-03-16 | 2018-09-04 | 南京邮电大学 | The identification in real time of single feature towards large-scale live scene and automatic screenshot method |
| CN108491822A (en) * | 2018-04-02 | 2018-09-04 | 杭州高创电子科技有限公司 | A kind of Face datection De-weight method based on the limited caching of embedded device |
| CN108513068A (en) * | 2018-03-30 | 2018-09-07 | 广东欧珀移动通信有限公司 | Choosing method, device, storage medium and the electronic equipment of image |
| CN109035246A (en) * | 2018-08-22 | 2018-12-18 | 浙江大华技术股份有限公司 | A kind of image-selecting method and device of face |
| CN109034013A (en) * | 2018-07-10 | 2018-12-18 | 腾讯科技（深圳）有限公司 | A kind of facial image recognition method, device and storage medium |
| CN109389019A (en) * | 2017-08-14 | 2019-02-26 | 杭州海康威视数字技术股份有限公司 | Facial image selection method, device and computer equipment |
| US10255668B2 (en) | 2016-08-23 | 2019-04-09 | International Business Machines Corporation | Enhancement of a profile photo system |
| CN109919091A (en) * | 2019-03-06 | 2019-06-21 | 广州佳都数据服务有限公司 | Face safety inspection method, device and electronic equipment based on dynamic white list |
| CN110059576A (en) * | 2019-03-26 | 2019-07-26 | 北京字节跳动网络技术有限公司 | Screening technique, device and the electronic equipment of picture |
| US10416859B2 (en) | 2016-08-23 | 2019-09-17 | International Business Machines Corporation | Enhanced configuration of a profile photo system |
| CN110706256A (en) * | 2019-09-27 | 2020-01-17 | 杭州博雅鸿图视频技术有限公司 | Detection tracking algorithm optimization method based on multi-core heterogeneous platform |
| CN110765913A (en) * | 2019-10-15 | 2020-02-07 | 浙江大华技术股份有限公司 | Human body target optimization method and device based on multiple evaluation indexes and storage medium |
| WO2020087434A1 (en) * | 2018-11-01 | 2020-05-07 | 深圳技术大学（筹） | Method and device for evaluating resolution of face image |
| CN111241925A (en) * | 2019-12-30 | 2020-06-05 | 新大陆数字技术股份有限公司 | Face quality evaluation method, system, electronic equipment and readable storage medium |
| WO2020114105A1 (en) * | 2018-12-03 | 2020-06-11 | 阿里巴巴集团控股有限公司 | Comparison method based on multiple facial images, apparatus and electronic device |
| CN111277746A (en) * | 2018-12-05 | 2020-06-12 | 杭州海康威视系统技术有限公司 | Indoor face snapshot method and system |
| CN111696128A (en) * | 2020-05-27 | 2020-09-22 | 南京博雅集智智能技术有限公司 | High-speed multi-target detection tracking and target image optimization method and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1731418A (en) * | 2005-08-19 | 2006-02-08 | 清华大学 | Method of robust accurate eye positioning in complicated background image |
| CN101059836A (en) * | 2007-06-01 | 2007-10-24 | 华南理工大学 | Human eye positioning and human eye state recognition method |
| CN101561710A (en) * | 2009-05-19 | 2009-10-21 | 重庆大学 | Man-machine interaction method based on estimation of human face posture |
| CN202197300U (en) * | 2010-08-05 | 2012-04-18 | 北京海鑫智圣技术有限公司 | Mobile face identification system |
| CN102799877A (en) * | 2012-09-11 | 2012-11-28 | 上海中原电子技术工程有限公司 | Method and system for screening face images |
-
2013
- 2013-12-27 CN CN201310737499.XA patent/CN103942525A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1731418A (en) * | 2005-08-19 | 2006-02-08 | 清华大学 | Method of robust accurate eye positioning in complicated background image |
| CN101059836A (en) * | 2007-06-01 | 2007-10-24 | 华南理工大学 | Human eye positioning and human eye state recognition method |
| CN101561710A (en) * | 2009-05-19 | 2009-10-21 | 重庆大学 | Man-machine interaction method based on estimation of human face posture |
| CN202197300U (en) * | 2010-08-05 | 2012-04-18 | 北京海鑫智圣技术有限公司 | Mobile face identification system |
| CN102799877A (en) * | 2012-09-11 | 2012-11-28 | 上海中原电子技术工程有限公司 | Method and system for screening face images |
Non-Patent Citations (1)
| Title |
|---|
| 龙剑锋: "近红外人脸图像识别与质量研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016164720A1 (en) * | 2015-04-10 | 2016-10-13 | Microsoft Technology Licensing, Llc | Snapshot capture for a communication session |
| US9509741B2 (en) | 2015-04-10 | 2016-11-29 | Microsoft Technology Licensing, Llc | Snapshot capture for a communication session |
| CN105554348A (en) * | 2015-12-25 | 2016-05-04 | 北京奇虎科技有限公司 | Image display method and device based on video information |
| CN106295585A (en) * | 2016-08-16 | 2017-01-04 | 深圳云天励飞技术有限公司 | A kind of filtration system of selection taking into account real-time and face quality and system |
| US10255668B2 (en) | 2016-08-23 | 2019-04-09 | International Business Machines Corporation | Enhancement of a profile photo system |
| US10416859B2 (en) | 2016-08-23 | 2019-09-17 | International Business Machines Corporation | Enhanced configuration of a profile photo system |
| US11163429B2 (en) | 2016-08-23 | 2021-11-02 | International Business Machines Corporation | Enhanced configuration of a profile photo system |
| CN106446851A (en) * | 2016-09-30 | 2017-02-22 | 厦门大图智能科技有限公司 | Visible light based human face optimal selection method and system |
| CN107358177A (en) * | 2017-06-27 | 2017-11-17 | 维拓智能科技（深圳）有限公司 | A kind of medium and long distance pedestrian detection method and terminal device based on graphical analysis |
| CN107346426B (en) * | 2017-07-10 | 2021-03-16 | 深圳市海清视讯科技有限公司 | Face information collection method based on camera face recognition |
| CN107346426A (en) * | 2017-07-10 | 2017-11-14 | 深圳市海清视讯科技有限公司 | A kind of face information collection method based on video camera recognition of face |
| CN109389019B (en) * | 2017-08-14 | 2021-11-05 | 杭州海康威视数字技术股份有限公司 | Face image selection method and device and computer equipment |
| CN109389019A (en) * | 2017-08-14 | 2019-02-26 | 杭州海康威视数字技术股份有限公司 | Facial image selection method, device and computer equipment |
| CN107633209A (en) * | 2017-08-17 | 2018-01-26 | 平安科技（深圳）有限公司 | Electronic installation, the method and storage medium of dynamic video recognition of face |
| WO2019033574A1 (en) * | 2017-08-17 | 2019-02-21 | 平安科技（深圳）有限公司 | Electronic device, dynamic video face recognition method and system, and storage medium |
| CN107958220A (en) * | 2017-12-06 | 2018-04-24 | 杭州魔点科技有限公司 | A kind of face database compression processing method and its intelligent apparatus based on recognition of face |
| CN108491784A (en) * | 2018-03-16 | 2018-09-04 | 南京邮电大学 | The identification in real time of single feature towards large-scale live scene and automatic screenshot method |
| CN108388885A (en) * | 2018-03-16 | 2018-08-10 | 南京邮电大学 | The identification in real time of more people's features towards large-scale live scene and automatic screenshot method |
| CN108491784B (en) * | 2018-03-16 | 2021-06-22 | 南京邮电大学 | Single person close-up real-time identification and automatic screenshot method for large live broadcast scene |
| CN108388885B (en) * | 2018-03-16 | 2021-06-08 | 南京邮电大学 | Multi-person close-up real-time identification and automatic screenshot method for large live broadcast scene |
| CN108513068A (en) * | 2018-03-30 | 2018-09-07 | 广东欧珀移动通信有限公司 | Choosing method, device, storage medium and the electronic equipment of image |
| CN108513068B (en) * | 2018-03-30 | 2021-03-02 | Oppo广东移动通信有限公司 | Image selection method and device, storage medium and electronic equipment |
| CN108491822B (en) * | 2018-04-02 | 2020-09-08 | 杭州高创电子科技有限公司 | Face detection duplication-removing method based on limited cache of embedded equipment |
| CN108491822A (en) * | 2018-04-02 | 2018-09-04 | 杭州高创电子科技有限公司 | A kind of Face datection De-weight method based on the limited caching of embedded device |
| CN109034013A (en) * | 2018-07-10 | 2018-12-18 | 腾讯科技（深圳）有限公司 | A kind of facial image recognition method, device and storage medium |
| US11557147B2 (en) | 2018-08-22 | 2023-01-17 | Zhejiang Dahua Technology Co., Ltd. | Systems and methods for selecting a best facial image of a target human face |
| CN109035246A (en) * | 2018-08-22 | 2018-12-18 | 浙江大华技术股份有限公司 | A kind of image-selecting method and device of face |
| CN109035246B (en) * | 2018-08-22 | 2020-08-04 | 浙江大华技术股份有限公司 | Face image selection method and device |
| WO2020087434A1 (en) * | 2018-11-01 | 2020-05-07 | 深圳技术大学（筹） | Method and device for evaluating resolution of face image |
| WO2020114105A1 (en) * | 2018-12-03 | 2020-06-11 | 阿里巴巴集团控股有限公司 | Comparison method based on multiple facial images, apparatus and electronic device |
| US11210502B2 (en) | 2018-12-03 | 2021-12-28 | Advanced New Technologies Co., Ltd. | Comparison method and apparatus based on a plurality of face image frames and electronic device |
| CN111277746A (en) * | 2018-12-05 | 2020-06-12 | 杭州海康威视系统技术有限公司 | Indoor face snapshot method and system |
| CN111277746B (en) * | 2018-12-05 | 2021-07-23 | 杭州海康威视系统技术有限公司 | Indoor face snapshot method and system |
| CN109919091A (en) * | 2019-03-06 | 2019-06-21 | 广州佳都数据服务有限公司 | Face safety inspection method, device and electronic equipment based on dynamic white list |
| CN110059576A (en) * | 2019-03-26 | 2019-07-26 | 北京字节跳动网络技术有限公司 | Screening technique, device and the electronic equipment of picture |
| CN110706256A (en) * | 2019-09-27 | 2020-01-17 | 杭州博雅鸿图视频技术有限公司 | Detection tracking algorithm optimization method based on multi-core heterogeneous platform |
| CN110765913A (en) * | 2019-10-15 | 2020-02-07 | 浙江大华技术股份有限公司 | Human body target optimization method and device based on multiple evaluation indexes and storage medium |
| CN111241925A (en) * | 2019-12-30 | 2020-06-05 | 新大陆数字技术股份有限公司 | Face quality evaluation method, system, electronic equipment and readable storage medium |
| CN111241925B (en) * | 2019-12-30 | 2023-08-18 | 新大陆数字技术股份有限公司 | Face quality assessment method, system, electronic equipment and readable storage medium |
| CN111696128A (en) * | 2020-05-27 | 2020-09-22 | 南京博雅集智智能技术有限公司 | High-speed multi-target detection tracking and target image optimization method and storage medium |
| CN111696128B (en) * | 2020-05-27 | 2024-03-12 | 南京博雅集智智能技术有限公司 | High-speed multi-target detection tracking and target image optimization method and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103942525A (en) | Real-time face optimal selection method based on video sequence | |
| CN104143079B (en) | The method and system of face character identification | |
| CN106570486B (en) | Filtered target tracking is closed based on the nuclear phase of Fusion Features and Bayes's classification | |
| CN103810490B (en) | A kind of method and apparatus for the attribute for determining facial image | |
| CN110532970B (en) | Age and gender attribute analysis method, system, equipment and medium for 2D images of human faces | |
| CN105740780B (en) | Method and device for detecting living human face | |
| CN103020992B (en) | A kind of video image conspicuousness detection method based on motion color-associations | |
| CN108549886A (en) | A kind of human face in-vivo detection method and device | |
| CN109670430A (en) | A kind of face vivo identification method of the multiple Classifiers Combination based on deep learning | |
| CN105989331B (en) | Face feature extraction element, facial feature extraction method, image processing equipment and image processing method | |
| CN107886507B (en) | A kind of salient region detecting method based on image background and spatial position | |
| CN104166841A (en) | Rapid detection identification method for specified pedestrian or vehicle in video monitoring network | |
| CN103473564B (en) | A kind of obverse face detection method based on sensitizing range | |
| CN105631455A (en) | Image main body extraction method and system | |
| CN103530599A (en) | Method and system for distinguishing real face and picture face | |
| CN105447859A (en) | Field wheat aphid counting method | |
| CN102609724B (en) | Method for prompting ambient environment information by using two cameras | |
| CN103440645A (en) | Target tracking algorithm based on self-adaptive particle filter and sparse representation | |
| CN105338343A (en) | No-reference stereo image quality evaluation method based on binocular perception | |
| CN111160291B (en) | Human eye detection method based on depth information and CNN | |
| CN105243376A (en) | Living body detection method and device | |
| CN106951826B (en) | Method for detecting human face and device | |
| CN109635634A (en) | A kind of pedestrian based on stochastic linear interpolation identifies data enhancement methods again | |
| CN108875474A (en) | Assess the method, apparatus and computer storage medium of face recognition algorithms | |
| CN108960142A (en) | Pedestrian based on global characteristics loss function recognition methods again |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140723 |
|
| RJ01 | Rejection of invention patent application after publication |