CN104616006A - Surveillance video oriented bearded face detection method - Google Patents
Surveillance video oriented bearded face detection method Download PDFInfo
- Publication number
- CN104616006A CN104616006A CN201510105012.5A CN201510105012A CN104616006A CN 104616006 A CN104616006 A CN 104616006A CN 201510105012 A CN201510105012 A CN 201510105012A CN 104616006 A CN104616006 A CN 104616006A
- Authority
- CN
- China
- Prior art keywords
- region
- face
- frame
- beard
- human body
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000010606 normalization Methods 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims description 19
- 210000000746 body region Anatomy 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 230000001815 facial effect Effects 0.000 claims description 3
- 238000005311 autocorrelation function Methods 0.000 claims description 2
- 230000007797 corrosion Effects 0.000 claims description 2
- 238000005260 corrosion Methods 0.000 claims description 2
- 238000005111 flow chemistry technique Methods 0.000 claims description 2
- 230000033001 locomotion Effects 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 4
- 230000000877 morphologic effect Effects 0.000 abstract description 2
- 238000011410 subtraction method Methods 0.000 abstract 1
- 230000006872 improvement Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000037237 body shape Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000036548 skin texture Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a surveillance video oriented bearded face detection method. The method mainly used for processing the surveillance videos and comprises the following steps: reading a video file into a computer, decoding, and obtaining a region with human upper body moving targets by using an improved background subtraction method and a human upper body HOG classifier; then, aiming at the upper one third part of the human upper body moving region, detecting the possible face region in the part by adopting a face Haar classifier, and intercepting the lower half part of the face region as a beard detection region; finally, carrying out scale normalization for the region, then calculating the horizontal texture complexity, and judging whether beards exist or not. The method utilizes the motion feature and the morphological feature of the surveillance video oriented bearded faces, can reliably detect the bearded face targets in the walking and running front upright bodies.
Description
Technical field
The invention belongs to the field of video image processing towards public safety prewarning, be specifically related to a kind of beard method for detecting human face.
Background technology
Video monitoring is widely used in public safety field, for the early warning in management of public safety business and verification provide strong data and technical support.Beard Face datection is mainly used in the bearded personnel targets of rapid screening face from magnanimity monitor video, public security department can be strengthened to the screening of specific objective, search efficiency, prevention and strike are broken laws and commit crime, trace suspicion personnel, safeguarded that social safety etc. has vital role.But also lower towards the intellectual analysis level of monitor video at present, in the beard Face datection of monitor video, there is no specially for the effective technology means of this application.Beard Face datection towards monitor video is the critical function of monitor video image procossing.Its treatment scheme is: first from monitor video, obtain view data, then extracts moving region, carries out Face datection further, finally judges whether face has beard.
For the links in beard Face datection, in moving object detection, existing method such as patent 201410110812.1 adopts ViBe algorithm to be that video frame image sets up background model, merge frame difference method segmentation foreground area, the method context update link is consuming time more, causes bulk treatment speed slower; Patent 201110253323.8 adopts carries out motion detection based on rim detection and frame difference method, and patent 201310586151.5 realizes moving target in conjunction with neighbor frame difference method and mixed Gauss model, and the deficiency of said method easily occurs hole region; In human detection, patent 201010218630.8 adopts the template detection multi-pose human body with ambiguity, and speed is slower; Patent 201310415544.X realizes human detection based on colored with depth information, but in monitor video, colouring information is larger by illumination effect; Patent 201110026465.0 carries out human detection based on depth image, is not suitable for conventional monitor video image; In face skin texture detection, patent 201210351040.1 utilizes Gabor filter to extract the textural characteristics of face, but only extracts facial overall profile, can not be used for representing beard feature; In addition the publication detected for face beard is not retrieved.
Summary of the invention
The present invention relates to a kind of beard method for detecting human face towards monitor video, the method processes mainly for monitor video, first on improvement Background difference basis, merge characteristics of human body and realize movement human region detection, then in conjunction with proportionate relationship and Haar classifier locating human face beard growth region, finally face beard situation is differentiated based on cross grain complexity.
The present invention can have the personnel targets of whiskers by rapid screening face from magnanimity monitor video, contribute to public security department screen better specific objective and search, improve the robotization of existing video monitoring system, intelligence degree, enhance public security quality monitoring.The method itself processes based on existing video completely, and without the need to additional triggers hardware, applied widely, intelligence degree is high.
Below the technical scheme in the present invention is described below:
1, the improvement Background difference merging characteristics of human body realizes movement human region detection
Background difference is a kind of method for testing motion of classics, has and calculates the advantage convenient, hole region is few.HOG sorter combines HOG characteristic sum Adaboost sorter, is often used to extract trunk feature and classify.The deficiency of existing Background difference is in background refresh process, easily cause indivedual background area to be mistaken for moving target.In the present invention, object is from video frame images, detect beard human face target, and this target belongs to a part for upper half of human body, and upper half of human body is included in moving region.Based on this feature, the present invention carries out detecting based on the upper half of human body of HOG sorter from the moving region that simple background subtraction value detects, be effective exercise region when upper half of human body being detected by current kinetic regional determination, context update is carried out to present frame, thus realizes the improvement to Background difference.Its specific implementation step is:
Step1: gather a large amount of upper half of human body area image in advance as positive sample, and gather a large amount of without upper half of human body area image as negative sample, in conjunction with positive and negative sample training upper half of human body HOG sorter;
Step2: choosing start frame image is initial background frame
, wherein
represent pixel coordinate in this frame;
Step3: obtain next frame successively, makes it be
,
for frame process sequence number,
for pixel coordinate, present frame and background frames are carried out the computing of difference binaryzation:
Wherein
for the error image after binaryzation,
for frame process sequence number,
for pixel coordinate,
for binary-state threshold, manually dynamically can set, generally can be set to
gray average;
Step4: the morphology operations current binaryzation error image first being corroded to rear expansion:
Wherein
for morphology template,
for the bianry image after corrosion,
for the bianry image after expansion,
for frame process sequence number,
for pixel coordinate;
Step5: right
extract the block sequence that non-zero pixels point is wherein formed
, wherein
for each block, add up to n.Travel through this block sequence, if the pixel count of each block
, then present frame is not containing the moving region met the demands, and obtains next frame, order
,
, go to Step3; If there is pixel count
block, then go to step6, wherein
for block of pixel amount threshold, make the height of present frame, to be widely respectively
,
,
can value be
;
Step6: will the block of block of pixel amount threshold condition be met
input human body HOG sorter, it exports as whether there is upper half of human body target in this block, and as existed, then this block of present frame is the moving region met the demands, and after having detected all blocks of present frame similarly, present frame is updated to background frames, namely
, then obtain next frame, order
, go to Step3; If all blocks of present frame all do not detect upper half of human body target, then present frame is without the moving region met the demands, and obtains next frame, order
, background frames does not upgrade, namely
, go to Step3;
By above-mentioned flow processing, until all frames are disposed.
2, the face beard region in conjunction with proportionate relationship and Haar classifier is located
Haar classifier has merged Haar-like characteristic sum Adaboost sorter, is often used to detection and location human face region.Existing method detects whole two field picture, is subject to the intensity profile approximate region interference in extraneous background, causes Face datection accuracy rate lower.In monitor video image, face is positioned at the top area of upper half of human body, and beard part is positioned at the latter half region of face, based on this feature, the present invention makes full use of proportionate relationship, adopt Haar classifier to detect in previous step and the top area of the upper half of human body navigated to carries out Face datection, reduce hunting zone, reduction background interference; Then extract its latter half from the human face region detected, be the region that pending beard detects.Concrete implementation step is:
Step1: gather a large amount of human face region image in advance as positive sample, and gather a large amount of without face area image as negative sample, in conjunction with positive and negative sample training face Haar classifier;
Step2: make us the upper half of human body area image called after that body HOG detection of classifier arrives
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the upper half of human body region sequence number in this frame,
for pixel coordinate, then the region of pending Face datection is
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the upper half of human body region sequence number in this frame,
for pixel coordinate;
Step3: will
be input to face Haar classifier, whether Output rusults is exists face, if there is face, and its region called after
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the human face region sequence number in this frame,
for pixel coordinate;
Step4: the region obtaining beard to be detected from human face region is
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the upper half of human body region sequence number in this frame,
for pixel coordinate.
3, the face beard based on cross grain complexity detects
What face beard reflected is local grain characteristic, the foundation that therefore can detect as beard with Texture complication.Conventional method needs the Texture complication calculating horizontal and vertical both direction.The present invention utilizes the directivity characteristics of beard growth, and namely beard gravitate has the trend naturally drooped usually, and its texture difference is mainly reflected in transverse direction, therefore only calculates its cross grain complexity.Concrete implementation step is:
Step1: for be detected beard region, be normalized to gray level image
, its width is 128, is highly 64,
for pixel coordinate, and
,
;
Step2: calculate
transverse autocorrelation function be:
Wherein
, and work as
time, order
;
Step3: constructing horizontal auto-correlation measure function is:
If
, then judge that current region is as beard face, otherwise be without beard face.Wherein
for facial roughness threshold value, beard can be had and without beard state face sample according to known, each the horizontal auto-correlation measure value calculated as stated above its corresponding region respectively
, average conduct is got to it
.
The invention has the advantages that:
1, the improvement Background difference merging characteristics of human body realizes movement human method for detecting area
(1) be effective exercise region by the regional determination containing upper half of human body image, effectively reduce the situation that background is erroneously detected as moving target;
(2) only just background is upgraded when detecting and there is upper half of human body moving region, can speed up processing and improve the moving target recognition quality of Background difference further.
2, in conjunction with the face beard region localization method of proportionate relationship and Haar classifier
(1) because upper half of human body shape facility has more typicalness than face characteristic, the accuracy of therefore searching for face in upper half of human body region is better than adopting single face Haar classifier to search for whole moving region;
(2) reduce face scope to be searched further in conjunction with proportionate relationship, thus improve bulk treatment speed and accuracy.
3, based on the face beard detection method of cross grain complexity
(1) adopt Texture complication to represent the degree of roughness of face, so as whether the bearded foundation of tool, simple, intuitive;
(2) only calculate cross grain complexity, more meet the face beard regularity of distribution, and reduce calculated amount, improve processing speed.
Accompanying drawing explanation
Fig. 1 is the overall schematic of the embodiment of the present invention;
Fig. 2 the present invention is based on to improve the schematic diagram that Background difference realizes upper half of human body moving overset grids.
Embodiment
Below in conjunction with diagram, the preferred embodiments of the present invention are described in detail.
As shown in Figure 1, first computing machine reads in video file to human body target testing flow process of the present invention, decodes and utilizes the Background difference improved in conjunction with upper half of human body HOG sorter, obtaining the region that there is upper half of human body moving target; Then for upper three/part of upper half of human body moving region, adopt face Haar classifier to detect the human face region that wherein may exist, and the latter half intercepted wherein is as beard surveyed area; Calculate its cross grain complexity after finally dimension normalization being carried out to this region, judge whether beard.The method utilizes motion feature and the morphological feature of beard face in monitor video, reliably can detect the beard human face target had in walking, the frontal upright human body of running.
Should be understood that, for those of ordinary skills, can be improved according to the above description or convert, such as change application etc., and all these improve and convert the protection domain that all should belong to claims of the present invention.
Technical scheme in the embodiment of the present invention is clearly and completely described, and obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Claims (4)
1. the beard method for detecting human face towards monitor video, process for monitor video, it is characterized in that, first computing machine reads in video file, decode and utilize Background difference in conjunction with upper half of human body HOG sorter, obtaining the region that there is upper half of human body moving target; Then for upper three/part of upper half of human body moving region, adopt face Haar classifier to detect the human face region that wherein may exist, and the latter half intercepted wherein is as beard surveyed area; Calculate its cross grain complexity after finally dimension normalization being carried out to this region, judge whether beard.
2. a kind of beard method for detecting human face towards monitor video according to claim 1, it is characterized in that, there is the region of upper half of human body moving target in described acquisition, concrete steps are as follows:
Step1: gather a large amount of upper half of human body area image in advance as positive sample, and gather a large amount of without upper half of human body area image as negative sample, in conjunction with positive and negative sample training upper half of human body HOG sorter;
Step2: choosing start frame image is initial background frame
, wherein
represent pixel coordinate in this frame;
Step3: obtain next frame successively, makes it be
,
for frame process sequence number,
for pixel coordinate, present frame and background frames are carried out the computing of difference binaryzation:
Wherein
for the error image after binaryzation,
for frame process sequence number,
for binary-state threshold, manually dynamically can set, generally can be set to
gray average;
Step4: the morphology operations current binaryzation error image first being corroded to rear expansion:
Wherein
for morphology template,
for the bianry image after corrosion,
for the bianry image after expansion,
for frame process sequence number;
Step5: right
extract the block sequence that non-zero pixels point is wherein formed
, wherein
for each block, add up to n, travel through this block sequence, if the pixel count of each block
, then present frame is not containing the moving region met the demands, and obtains next frame, order
,
, go to Step3; If there is pixel count
block, then go to step6, wherein
for block of pixel amount threshold, make the height of present frame, to be widely respectively
,
,
can value be
;
Step6: will the block of block of pixel amount threshold condition be met
input human body HOG sorter, it exports as whether there is upper half of human body target in this block, and as existed, then this block of present frame is the moving region met the demands, and after having detected all blocks of present frame similarly, present frame is updated to background frames, namely
, then obtain next frame, order
, go to Step3; If all blocks of present frame all do not detect upper half of human body target, then present frame is the moving region without meeting the demands, and obtains next frame, order
, background frames does not upgrade, namely
, go to Step3;
By above-mentioned flow processing, until all frames are disposed.
3. a kind of beard method for detecting human face towards monitor video according to claim 1, it is characterized in that, described intercepting the latter half is wherein as beard surveyed area, and concrete steps are as follows:
Step3.1: gather a large amount of human face region image in advance as positive sample, and gather a large amount of without face area image as negative sample, in conjunction with positive and negative sample training face Haar classifier;
Step3.2: make us the upper half of human body area image called after that body HOG detection of classifier arrives
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the upper half of human body region sequence number in this frame, then the region of pending Face datection is
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the upper half of human body region sequence number in this frame;
Step3.3: will
be input to face Haar classifier, whether Output rusults is exists face, if there is face, and its region called after
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the human face region sequence number in this frame;
Step3.4: the region obtaining beard to be detected from human face region is
, its top left corner pixel point coordinate is
, this region is high, wide to be respectively
,
, wherein
for frame process sequence number,
for the upper half of human body region sequence number in this frame.
4. a kind of beard method for detecting human face towards monitor video according to claim 1, is characterized in that, described dimension normalization is carried out to region after calculate its cross grain complexity, judged whether beard, concrete steps are as follows:
Step4.1: for be detected beard region, be normalized to gray level image
, its width is 128, is highly 64,
for pixel coordinate, and
,
;
Step4.2: calculate
transverse autocorrelation function be:
Wherein
, and work as
time, order
;
Step4.3: constructing horizontal auto-correlation measure function is:
If
, then judge that current region is as beard face, otherwise be without beard face, wherein
for facial roughness threshold value, there is beard and without beard state face sample according to known, each the horizontal auto-correlation measure value calculated as stated above its corresponding region respectively
, average conduct is got to it
.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510105012.5A CN104616006B (en) | 2015-03-11 | 2015-03-11 | A kind of beard method for detecting human face towards monitor video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510105012.5A CN104616006B (en) | 2015-03-11 | 2015-03-11 | A kind of beard method for detecting human face towards monitor video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104616006A true CN104616006A (en) | 2015-05-13 |
| CN104616006B CN104616006B (en) | 2017-08-25 |
Family
ID=53150443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510105012.5A Expired - Fee Related CN104616006B (en) | 2015-03-11 | 2015-03-11 | A kind of beard method for detecting human face towards monitor video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104616006B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104866843A (en) * | 2015-06-05 | 2015-08-26 | 中国人民解放军国防科学技术大学 | Monitoring-video-oriented masked face detection method |
| CN105975952A (en) * | 2016-05-26 | 2016-09-28 | 天津艾思科尔科技有限公司 | Beard detection method and system in video image |
| CN106295524A (en) * | 2016-08-01 | 2017-01-04 | 马平 | A kind of human motion recognition method of view-based access control model word bag |
| CN107346417A (en) * | 2017-06-13 | 2017-11-14 | 浪潮金融信息技术有限公司 | Method for detecting human face and device |
| CN108109115A (en) * | 2017-12-07 | 2018-06-01 | 深圳大学 | Enhancement Method, device, equipment and the storage medium of character image |
| CN108197507A (en) * | 2017-12-30 | 2018-06-22 | 刘智 | A kind of privacy real-time protection method and system |
| CN108764185A (en) * | 2018-06-01 | 2018-11-06 | 京东方科技集团股份有限公司 | A kind of image processing method and device |
| CN108932465A (en) * | 2017-12-28 | 2018-12-04 | 浙江宇视科技有限公司 | Reduce the method, apparatus and electronic equipment of Face datection false detection rate |
| CN111027474A (en) * | 2019-12-09 | 2020-04-17 | Oppo广东移动通信有限公司 | Face area acquisition method and device, terminal equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024145A (en) * | 2010-12-01 | 2011-04-20 | 五邑大学 | Layered recognition method and system for disguised face |
-
2015
- 2015-03-11 CN CN201510105012.5A patent/CN104616006B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024145A (en) * | 2010-12-01 | 2011-04-20 | 五邑大学 | Layered recognition method and system for disguised face |
Non-Patent Citations (3)
| Title |
|---|
| JIAN-GANG WANG ETC.: ""Real-time moustache detection by combining image decolorization and texture detection with applications to facial gender recognition"", 《MACHINE VISION AND APPLICATIONS》 * |
| 王喆: ""面向自动柜员机智能安防的异常人脸检测技术和系统研发"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 练士龙等: ""视频监控系统中的行人及其面部侦测研究"", 《电子设计工程》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104866843B (en) * | 2015-06-05 | 2018-08-21 | 中国人民解放军国防科学技术大学 | A kind of masked method for detecting human face towards monitor video |
| CN104866843A (en) * | 2015-06-05 | 2015-08-26 | 中国人民解放军国防科学技术大学 | Monitoring-video-oriented masked face detection method |
| CN105975952A (en) * | 2016-05-26 | 2016-09-28 | 天津艾思科尔科技有限公司 | Beard detection method and system in video image |
| CN106295524A (en) * | 2016-08-01 | 2017-01-04 | 马平 | A kind of human motion recognition method of view-based access control model word bag |
| CN107346417A (en) * | 2017-06-13 | 2017-11-14 | 浪潮金融信息技术有限公司 | Method for detecting human face and device |
| CN107346417B (en) * | 2017-06-13 | 2020-09-01 | 浪潮金融信息技术有限公司 | Face detection method and device |
| CN108109115A (en) * | 2017-12-07 | 2018-06-01 | 深圳大学 | Enhancement Method, device, equipment and the storage medium of character image |
| CN108932465A (en) * | 2017-12-28 | 2018-12-04 | 浙江宇视科技有限公司 | Reduce the method, apparatus and electronic equipment of Face datection false detection rate |
| CN108932465B (en) * | 2017-12-28 | 2021-02-02 | 浙江宇视科技有限公司 | Method and device for reducing false detection rate of face detection and electronic equipment |
| CN108197507A (en) * | 2017-12-30 | 2018-06-22 | 刘智 | A kind of privacy real-time protection method and system |
| CN108764185A (en) * | 2018-06-01 | 2018-11-06 | 京东方科技集团股份有限公司 | A kind of image processing method and device |
| US11321952B2 (en) | 2018-06-01 | 2022-05-03 | Boe Technology Group Co., Ltd. | Computer-implemented method of alerting driver of vehicle, apparatus for alerting driver of vehicle, vehicle, and computer-program product |
| CN111027474A (en) * | 2019-12-09 | 2020-04-17 | Oppo广东移动通信有限公司 | Face area acquisition method and device, terminal equipment and storage medium |
| CN111027474B (en) * | 2019-12-09 | 2024-03-15 | Oppo广东移动通信有限公司 | Face region acquisition method and device, terminal equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104616006B (en) | 2017-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104616006A (en) | Surveillance video oriented bearded face detection method | |
| CN107330920B (en) | Monitoring video multi-target tracking method based on deep learning | |
| Stalder et al. | Cascaded confidence filtering for improved tracking-by-detection | |
| CN101739686B (en) | Moving object tracking method and system thereof | |
| CN102509078B (en) | Fire detection device based on video analysis | |
| CN104978567B (en) | Vehicle checking method based on scene classification | |
| CN103246896B (en) | A kind of real-time detection and tracking method of robustness vehicle | |
| CN102542289A (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| CN105260749B (en) | Real-time target detection method based on direction gradient binary pattern and soft cascade SVM | |
| CN105469105A (en) | Cigarette smoke detection method based on video monitoring | |
| CN104866843B (en) | A kind of masked method for detecting human face towards monitor video | |
| CN103824070A (en) | Rapid pedestrian detection method based on computer vision | |
| CN105893946A (en) | Front face image detection method | |
| Szwoch | Extraction of stable foreground image regions for unattended luggage detection | |
| CN102147861A (en) | Moving target detection method for carrying out Bayes judgment based on color-texture dual characteristic vectors | |
| CN104732220A (en) | Specific color human body detection method oriented to surveillance videos | |
| CN105513053A (en) | Background modeling method for video analysis | |
| CN105554462A (en) | Remnant detection method | |
| Ling et al. | A background modeling and foreground segmentation approach based on the feedback of moving objects in traffic surveillance systems | |
| CN107103303A (en) | A kind of pedestrian detection method based on GMM backgrounds difference and union feature | |
| Yadav | Efficient method for moving object detection in cluttered background using Gaussian Mixture Model | |
| Bullkich et al. | Moving shadow detection by nonlinear tone-mapping | |
| CN104899559B (en) | A kind of rapid pedestrian detection method based on video monitoring | |
| CN103577832A (en) | People flow statistical method based on spatio-temporal context | |
| Zhao et al. | APPOS: An adaptive partial occlusion segmentation method for multiple vehicles tracking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170825 |