CN106056079A - Image acquisition device and facial feature occlusion detection method - Google Patents
Image acquisition device and facial feature occlusion detection method Download PDFInfo
- Publication number
- CN106056079A CN106056079A CN201610375049.4A CN201610375049A CN106056079A CN 106056079 A CN106056079 A CN 106056079A CN 201610375049 A CN201610375049 A CN 201610375049A CN 106056079 A CN106056079 A CN 106056079A
- Authority
- CN
- China
- Prior art keywords
- detection
- face
- image
- capture device
- blocking
- 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 84
- 230000001815 facial effect Effects 0.000 title abstract 3
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000012544 monitoring process Methods 0.000 claims abstract description 11
- 230000000903 blocking effect Effects 0.000 claims description 30
- 210000001508 eye Anatomy 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000006835 compression Effects 0.000 claims description 4
- 238000007906 compression Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000005252 bulbus oculi Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/285—Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system
-
- 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
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F19/00—Complete banking systems; Coded card-freed arrangements adapted for dispensing or receiving monies or the like and posting such transactions to existing accounts, e.g. automatic teller machines
- G07F19/20—Automatic teller machines [ATMs]
- G07F19/207—Surveillance aspects at ATMs
Landscapes
- Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Geometry (AREA)
- Accounting & Taxation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image acquisition device and a facial feature occlusion detection method. The method comprises steps: 1, the image acquisition device is used for acquiring a video image in a monitoring area; 2, background modeling and foreground detection are carried out on the video image acquired by the image acquisition device, and whether a human face appears in the video image is judged; 3, when a human face is detected to appear in the video image, whether occlusion happens to the image acquisition device is detected, and an alarming prompt is started in the case of occlusion, and a fourth step is carried out if no occlusion appears; and 4, occlusion detection on the human face and the facial features, that is, human face integrity detection, is continued, and thus, whether human face occlusion exists in the image acquisition device is judged. The method of the invention has the advantages that the real-time performance is good; the detection accuracy is high; an automatic and intelligent monitoring means is provided for an automatic teller machine system of a bank; and a technical support is provided for a bank unattended operation and management mode.
Description
Technical field
The present invention relates to field of video monitoring, a kind of image capture device and the occlusion detection side of human face five-sense-organ
Method.
Background technology
In recent years, the crime of ATM presents state occurred frequently, and the masked camouflage of offender, malice block camera lens etc.
Criminal activity happens occasionally, and the normal Financial Management order of serious harm, public safety problem based on ATM is subject to
Arrive increasing attention.But traditional monitoring mode is the mode using " only record do not judge ", can only incident it
By the playback of video abnormal conditions investigated afterwards and round, it is impossible to accomplishing to judge in real time and report to the police.Therefore, based on
The abnormal occlusion detection of monitor video, can realize Deviant Behavior in monitoring scene by the technology such as image procossing, machine learning
Automatically detection and alert operation, be possible not only to greatly reduce human cost, the process of monitoring system can also be improved simultaneously
Ability, has the most wide application prospect.
The present invention images first-class image capture device and human face five-sense-organ occlusion detection method, utilizes computer to be automatically performed figure
Block as collecting device and detection that human face five-sense-organ is blocked, determine whether the appearance of face by extracting target prospect,
Judging have face, after occurring, image is carried out feature extraction, carrying out the detection that image capture device blocks, re-use grader
Realize detection that face and face are blocked, thus realize that accuracy is higher, the preferable detection method of real-time.
Summary of the invention
It is an object of the invention to provide the image capture device in a kind of video monitoring system and human face five-sense-organ blocks inspection
Survey method so that computer can be automatically performed the occlusion detection to image capture device and human face five-sense-organ, ensures higher simultaneously
Accuracy, and preferably real-time.
For reaching object above, a kind of image capture device of present invention offer and the occlusion detection method bag of human face five-sense-organ
Include following steps:
Step 1: utilize the video image in image capture device acquisition monitoring region；
Step 2: the video image collected for described image capture device carries out background modeling and detection prospect,
And judge in described video image the appearance with or without face；
Step 3: when detect there is face in described video image time, detect described image capture device whether occur hide
Gear, starts alarm when judging to occur blocking, and when judging not occur blocking, goes to step 4；
Step 4: proceed the occlusion detection of face and face, i.e. face integrity detection, and judge that image is adopted accordingly
Blocking of face whether is there is in collection equipment.
Alternatively, described step 1 also includes the step processing and displaying the image collected.
Alternatively, the video image utilizing mixed Gaussian method to collect for described image capture device carries out background
Modeling.
Alternatively, described step 2 further includes steps of
Step 21, sets up gauss hybrid models；
Step 22, utilized the parameter value in a upper moment to be updated for the parameter of current time gauss hybrid models；
Step 23, the mixed Gauss model after utilizing parameter to update carrys out the feature of each pixel in phenogram picture, and according to
This carries out foreground detection, obtains foreground image；
Step 24, what the foreground image obtaining described step 23 carried out morphologic filtering opens operation；
Step 25, utilize connected domain analysis method judge after described step 24 processes in the foreground image that obtains with or without
The appearance of face.
Alternatively, in described step 23, when carrying out foreground detection, certain pixel of present image is mixed with described
The pixel of corresponding position that Gauss model characterizes mates, if the match is successful, judge this pixel as background dot,
It it is otherwise foreground point.
Alternatively, described step 25 further includes steps of
Step 251, utilizes connected domain analysis method to carry out for the foreground image obtained after described step 24 processes
Segmentation, obtains multiple connected domain；
Step 252, calculates the profile length of each connected domain that described step 251 obtains, when there is a connected domain
When profile length is more than a certain default minimum threshold, represents in image capture device and have face to occur.
Alternatively, described step 3 further includes steps of
Step 31, carries out gray processing for described video image, obtains gray level image；
Step 32, carries out the extraction of Sobel gradient information to the gray level image obtained；
Step 33, carries out binarization operation to extracting the gradient information obtained；
Step 34, statistical gradient value is more than the number of the pixel of a certain predetermined threshold value, when the pixel number obtained surpasses
When crossing some, i.e. think that blocking occurs in image capture device；
Step 35, obtains carrying out alarm, the most again when blocking occurs in image capture device in the detection of described step 34
Proceed the detection of next frame sequence of video images, go out when being consecutively detected image capture device in a preset time threshold
When now blocking, then carry out the alarm of refusal service, terminate the work of occlusion detection simultaneously.
Alternatively, described step 4 further includes steps of
Step 41, is compressed described video image；
Step 42, carries out histogram equalization operation to the image after compression；
Step 43, carries out the detection of face in image；
Step 44, when detecting the presence of face, obtain the left upper apex coordinate of face place boundary rectangle frame and width,
Height, judges that the size of boundary rectangle frame whether in certain threshold range, the most then detects face correct simultaneously, otherwise it is assumed that
The face mistake of detection；
Step 45, in the boundary rectangle frame that Face datection obtains, the sense being respectively provided with eyes, nose and mouth detection is emerging
Interest region, then carry out the detection of eye, nose and mouth the most respectively, when face detection is complete, i.e. illustrate
Blocking of face does not occur, otherwise it is assumed that blocking and carrying out alarm of face occurs in image capture device；
Step 46, carries out alarm what face occurred in detection, is then further continued for carrying out next frame video figure when blocking
As the detection of sequence, when being consecutively detected in a preset time threshold, when blocking of human face five-sense-organ occurs, then carry out refusal clothes
The alarm of business, the work of termination detection simultaneously.
Alternatively, in described step 43, Adaboost cascade classifier based on Haar-like feature is used to carry out image
The detection of middle face.
Alternatively, in described step 45, Adaboost cascade classifier based on Haar-like is used to carry out human eye respectively
The detection of eyeball, nose and mouth.
The method that in the present invention, image capture device and human face five-sense-organ occlusion detection are used have higher accuracy with
And preferable real-time, the platform such as automated teller machine platform for banking system provides automatization, intelligentized monitoring means,
Can promote really to realize the abnormal detection blocked and alert operation, the non-attendant operation for platforms such as ATMs manages
Pattern provides strong technical support.
Accompanying drawing explanation
Fig. 1 is image capture device and the flow chart of human face five-sense-organ occlusion detection method according to an embodiment of the invention.
Fig. 2 is the flow chart of foreground extracting method according to an embodiment of the invention.
Fig. 3 (a), (b) are respectively in one embodiment of the invention the Sobel warp factor on the X and Y-direction related to.
Fig. 4 (a) is to the edge feature schematic diagram in image zooming-out Haar-like feature in one embodiment of the invention.
Fig. 4 (b) is to the linear character schematic diagram in image zooming-out Haar-like feature in one embodiment of the invention.
Fig. 4 (c) is to the center ring characteristics signal in image zooming-out Haar-like feature in one embodiment of the invention
Figure.
Fig. 5 is the schematic diagram of the integrogram that one embodiment of the invention relates to.
Fig. 6 is the method schematic diagram using integrogram to calculate feature in one embodiment of the invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment, and reference
Accompanying drawing, the present invention is described in more detail.
Fig. 1 is image capture device and the flow chart of human face five-sense-organ occlusion detection method according to an embodiment of the invention,
As it is shown in figure 1, described image capture device and human face five-sense-organ occlusion detection method comprise the following steps:
Step 1: utilize the video image in image capture device acquisition monitoring region；
Wherein, described image capture device can be digital image acquisition apparatus, such as digital camera, it is also possible to for mould
Intend the image capture devices such as image capture device.
In an embodiment of the present invention, described step 1 also includes the figure utilizing the electronic equipments such as computer to collecting
As the step processed and displayed, wherein, the process carrying out image includes binary conversion treatment and morphologic filtering operation.
Step 2: the video image collected for described image capture device carries out background modeling and detection prospect,
And judge in described video image the appearance with or without face；
In an embodiment of the present invention, the video that mixed Gaussian method collects is utilized for described image capture device
Image carries out background modeling, naturally it is also possible to use other background modeling methods, is not especially limited its present invention.
Further, as in figure 2 it is shown, described utilize mixed Gaussian method to collect for described image capture device
Image carries out background modeling and detection prospect, and judges in described image that the step of the appearance with or without face includes following step
Rapid:
Step 21, sets up gauss hybrid models；
In this step, K Gauss model is set up for each pixel in image, wherein, for certain pixel of t
Sample value xt, its probability density function is represented by the probability density function weighted sum of K Multi-dimensional Gaussian distribution function, as
Shown in formula (1):
In formula, K is the number of Gauss model, ωI, tFor Gauss model weights, μI, tFor the average of i-th Gauss distribution,For covariance matrix, whereinFor this pixel in the variance of the i-th Gauss model of t image sequence,
ηI, t(xt, μI, t, ∑I, t) it is the i-th Gauss distribution of t, as shown in formula (2):
Wherein, n is xtDimension.
Step 22, utilized the parameter value in a upper moment to be updated for the parameter of current time gauss hybrid models；
In this step, the meansigma methods of i-th Gauss distribution and the parameter of variance are come more respectively shown in formula (3), (4)
New:
Wherein,For the renewal speed of learning rate, i.e. model, μI, t-1Represented i-th Gauss distribution average in a upper moment
Value, μI, tRepresent the meansigma methods of the i-th Gauss distribution of current time,Represented the side of the i-th Gauss distribution in a upper moment
Difference,Represent the variance of the i-th Gauss distribution of current time.
Step 23, the mixed Gauss model after utilizing parameter to update carrys out the feature of each pixel in phenogram picture, and according to
This carries out foreground detection, obtains foreground image, specifically, when carrying out foreground detection, by certain pixel of present image
The average of the mixed Gauss model that pixel value is corresponding with this pixel compares, when both distances are less than this mixed Gauss model
3 times of variance time, represent that the match is successful, if the match is successful, judges that this pixel, as background dot, is otherwise foreground point；
Step 24, what the foreground image obtaining described step 23 carried out morphologic filtering opens operation, the most first corrodes the most swollen
Swollen, remove and foreground image can also improve while incoherent noise the integrity extracting image；
Step 25, utilize connected domain analysis method judge after described step 24 processes in the foreground image that obtains with or without
The appearance of face.
This step further includes steps of
Step 251, utilizes connected domain analysis method to carry out for the foreground image obtained after described step 24 processes
Segmentation, obtains multiple connected domain；
Step 252, calculates the profile length of each connected domain that described step 251 obtains, when there is a connected domain
When profile length is more than a certain default minimum threshold, represents in image capture device and have face to occur, such as, work as image resolution
When rate is 640 × 480, the threshold value of minimized profile length may be configured as 5000.
Step 3: when detect there is face in described video image time, detect described image capture device whether occur hide
Gear, starts alarm when judging to occur blocking, and when judging not occur blocking, goes to step 4；
In described step 3, when image capture device is carried out occlusion detection, first described video image is carried out gray processing
Obtain gray level image, then extract the Gradient Features information of gray level image, thus carry out the occlusion detection of image capture device, tool
Body ground, described step 3 comprises the following steps:
Step 31, carries out gray processing for described video image, obtains gray level image；
Step 32, carries out the extraction of Sobel gradient information, in an embodiment of the present invention, choosing to the gray level image obtained
The warp factor taken is as it is shown on figure 3, in Fig. 3, (a) (b) figure represents the derivative calculating X and Y-direction respectively；
Step 33, carries out binarization operation, if the pixel value of initial video image is to extracting the gradient information obtained
Pixel, the pixel value after binaryzation is R, then shown in binarization such as formula (5):
Wherein, threshold represents binary-state threshold.
Step 34, statistical gradient value is more than the number of the pixel of a certain predetermined threshold value (such as can be set to 245～250),
When the pixel number obtained exceedes some, i.e. think that blocking occurs in image capture device；
Step 35, obtains carrying out alarm, the most again when blocking occurs in image capture device in the detection of described step 34
Proceed the detection of next frame sequence of video images, when being consecutively detected image in a preset time threshold (such as 5 seconds)
When blocking occurs in collecting device, then carry out the alarm of refusal service, terminate the work of occlusion detection simultaneously.
Step 4: the testing result obtained for described step 3 determines whether, when image capture device do not occur
When blocking, proceed the occlusion detection of face and face, i.e. face integrity detection, and judge image capture device accordingly
The most whether there is blocking of face.
In this step, sequence of video images is carried out the integrity detection of the face such as face and eyes, nose and mouth, tool
Body ground, described step 4 further includes steps of
Step 41, reduces by a certain percentage by described video image, thus improves the speed of detection；Wherein, the ratio of compression
Example scope can be taken as between 1.1～1.3.
Step 42, carries out histogram equalization operation to the image after compression, removes uncorrelated noise present in image；
Step 43, uses Adaboost cascade classifier based on Haar-like feature to carry out the detection of face in image；
In an embodiment of the present invention, as shown in Figure 4, wherein, figure (a) represents 14 kinds of Haar-like features of 3 class of use
The edge feature extracted, figure (b) represents that the linear character extracted, figure (c) represent the center ring characteristics extracted.Obtain in extraction
After detection feature, use integrogram as shown in Figure 5 that the calculating of eigenvalue is accelerated, i.e. each pixel in coordinate
The value of integrogram is its upper left corner all pixels sum, in calculating detection window after the integrogram of each pixel,
The value of the integrogram of detection window can be calculated as shown in Figure 6 shown in equation below (6):
S (D)=S (A+B+C+D)+S (A)-S (A+B)-S (A+C) (6)
Wherein, S () represents the size specifying region, and such as S (A) represents the size of a-quadrant, and a-quadrant
Scope as shown in Figure 6.
Step 44, when detecting the presence of face, obtain the left upper apex coordinate of face place boundary rectangle frame and width,
Height, judges that whether the size of boundary rectangle frame is in certain threshold value (between such as (220,220)～(280,280)) scope simultaneously
In, the most then detection face is correct, otherwise it is assumed that the face mistake of detection；
Step 45, Face datection obtain boundary rectangle frame in, selected part size, be respectively provided with eyes, nose with
And the area-of-interest of mouth detection, then Adaboost cascade classifier based on Haar-like is used to divide in the region of interest
Do not carry out the detection of eye, nose and mouth, when face detection is complete, i.e. in explanation image capture device, face does not occur
Block, otherwise it is assumed that blocking and carrying out alarm of face occurs；
Step 46, carries out alarm what face occurred in detection, is then further continued for carrying out next frame video figure when blocking
As the detection of sequence, when being consecutively detected in a preset time threshold (such as 5 seconds), when blocking of human face five-sense-organ occurs, then enter
The alarm of row refusal service, the work of termination detection simultaneously.
In sum, the present invention proposes a kind of occlusion detection method of image capture device and human face five-sense-organ.
Particular embodiments described above, has been carried out the purpose of the present invention, technical scheme and beneficial effect the most in detail
Describe in detail bright, be it should be understood that the specific embodiment that the foregoing is only the present invention, be not limited to the present invention, all
Within the spirit and principles in the present invention, any modification, equivalent substitution and improvement etc. done, should be included in the guarantor of the present invention
Within the scope of protecting.
Claims (10)
1. an image capture device and human face five-sense-organ occlusion detection method, it is characterised in that comprise the following steps:
Step 1: utilize the video image in image capture device acquisition monitoring region；
Step 2: the video image collected for described image capture device carries out background modeling and detection prospect, and sentences
With or without the appearance of face in disconnected described video image；
Step 3: when detect there is face in described video image time, detect whether described image capture device occurs blocking,
Start alarm when judging to occur blocking, when judging not occur blocking, go to step 4；
Step 4: proceed the occlusion detection of face and face, i.e. face integrity detection, and judge that image acquisition sets accordingly
Blocking of face whether is there is for interior.
Method the most according to claim 1, it is characterised in that described step 1 also includes carrying out the image collected
Process and the step of display.
Method the most according to claim 1, it is characterised in that utilize mixed Gaussian method for described image capture device
The video image collected carries out background modeling.
Method the most according to claim 1, it is characterised in that described step 2 further includes steps of
Step 21, sets up gauss hybrid models；
Step 22, utilized the parameter value in a upper moment to be updated for the parameter of current time gauss hybrid models；
Step 23, the mixed Gauss model after utilizing parameter to update carrys out the feature of each pixel in phenogram picture, and enters accordingly
Row foreground detection, obtains foreground image；
Step 24, what the foreground image obtaining described step 23 carried out morphologic filtering opens operation；
Step 25, utilizes connected domain analysis method to judge in the foreground image obtained after described step 24 processes with or without face
Appearance.
Method the most according to claim 4, it is characterised in that in described step 23, when carrying out foreground detection, will be current
The pixel of the corresponding position that certain pixel of image characterizes with described mixed Gauss model mates, if mated into
Merit then judges that this pixel, as background dot, is otherwise foreground point.
Method the most according to claim 4, it is characterised in that described step 25 further includes steps of
Step 251, utilizes connected domain analysis method to split for the foreground image obtained after described step 24 processes,
Obtain multiple connected domain；
Step 252, calculates the profile length of each connected domain that described step 251 obtains, when the profile that there is a connected domain
When length is more than a certain default minimum threshold, represents in image capture device and have face to occur.
Method the most according to claim 1, it is characterised in that described step 3 further includes steps of
Step 31, carries out gray processing for described video image, obtains gray level image；
Step 32, carries out the extraction of Sobel gradient information to the gray level image obtained；
Step 33, carries out binarization operation to extracting the gradient information obtained；
Step 34, statistical gradient value is more than the number of the pixel of a certain predetermined threshold value, when the pixel number obtained is more than one
During determined number, i.e. think that blocking occurs in image capture device；
Step 35, obtains carrying out alarm when blocking occurs in image capture device in the detection of described step 34, is then further continued for
Carry out the detection of next frame sequence of video images, occur hiding when being consecutively detected image capture device in a preset time threshold
During gear, then carry out the alarm of refusal service, terminate the work of occlusion detection simultaneously.
Method the most according to claim 1, it is characterised in that described step 4 further includes steps of
Step 41, is compressed described video image；
Step 42, carries out histogram equalization operation to the image after compression；
Step 43, carries out the detection of face in image；
Step 44, when detecting the presence of face, obtains left upper apex coordinate and width, the height of face place boundary rectangle frame,
Judge that the size of boundary rectangle frame whether in certain threshold range, the most then detects face correct simultaneously, otherwise it is assumed that inspection
The face mistake surveyed；
Step 45, in the boundary rectangle frame that Face datection obtains, is respectively provided with eyes, nose and the region of interest of mouth detection
Territory, then carry out the detection of eye, nose and mouth the most respectively, when face detection is complete, i.e. explanatory diagram picture
Blocking of face does not occur, otherwise it is assumed that blocking and carrying out alarm of face occurs in collecting device；
Step 46, carries out alarm what face occurred in detection, is then further continued for carrying out next frame video image sequence when blocking
, there is when blocking of human face five-sense-organ when being consecutively detected in a preset time threshold in the detection of row, then carry out refusal service
Alarm, the simultaneously work of termination detection.
Method the most according to claim 8, it is characterised in that in described step 43, uses based on Haar-like feature
Adaboost cascade classifier carries out the detection of face in image.
Method the most according to claim 1, it is characterised in that in described step 45, uses based on Haar-like
Adaboost cascade classifier carries out the detection of eye, nose and mouth respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610375049.4A CN106056079B (en) | 2016-05-31 | 2016-05-31 | A kind of occlusion detection method of image capture device and human face five-sense-organ |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610375049.4A CN106056079B (en) | 2016-05-31 | 2016-05-31 | A kind of occlusion detection method of image capture device and human face five-sense-organ |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106056079A true CN106056079A (en) | 2016-10-26 |
| CN106056079B CN106056079B (en) | 2019-07-05 |
Family
ID=57173186
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610375049.4A Active CN106056079B (en) | 2016-05-31 | 2016-05-31 | A kind of occlusion detection method of image capture device and human face five-sense-organ |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106056079B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107316312A (en) * | 2017-06-30 | 2017-11-03 | 深圳信路通智能技术有限公司 | A kind of video image occlusion detection method and system |
| CN107633204A (en) * | 2017-08-17 | 2018-01-26 | 平安科技（深圳）有限公司 | Face occlusion detection method, apparatus and storage medium |
| CN107909065A (en) * | 2017-12-29 | 2018-04-13 | 百度在线网络技术（北京）有限公司 | The method and device blocked for detecting face |
| CN107958572A (en) * | 2017-11-30 | 2018-04-24 | 西安科锐盛创新科技有限公司 | A kind of baby monitoring systems |
| CN108009496A (en) * | 2017-11-30 | 2018-05-08 | 西安科锐盛创新科技有限公司 | Face blocks recognition methods |
| CN108734927A (en) * | 2018-04-27 | 2018-11-02 | 济南中维世纪科技有限公司 | A kind of electronic peephole viewer blocks alarm system |
| CN109101923A (en) * | 2018-08-14 | 2018-12-28 | 罗普特（厦门）科技集团有限公司 | A kind of personnel wear the detection method and device of mask situation |
| CN110516649A (en) * | 2019-09-02 | 2019-11-29 | 南京微小宝信息技术有限公司 | Alumnus's authentication method and system based on recognition of face |
| CN110647858A (en) * | 2019-09-29 | 2020-01-03 | 上海依图网络科技有限公司 | Video occlusion judgment method and device and computer storage medium |
| CN111951176A (en) * | 2020-06-30 | 2020-11-17 | 重庆灵翎互娱科技有限公司 | Method and equipment for removing local shadows of face image |
| CN112487921A (en) * | 2020-11-25 | 2021-03-12 | 奥比中光科技集团股份有限公司 | Face image preprocessing method and system for living body detection |
| CN112651322A (en) * | 2020-12-22 | 2021-04-13 | 北京眼神智能科技有限公司 | Cheek shielding detection method and device and electronic equipment |
| CN112801963A (en) * | 2021-01-19 | 2021-05-14 | 湖南优美科技发展有限公司 | Video image occlusion detection method and system |
| CN114708657A (en) * | 2022-03-30 | 2022-07-05 | 深圳可视科技有限公司 | Student attention detection method and system based on multimedia teaching |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080247609A1 (en) * | 2007-04-06 | 2008-10-09 | Rogerio Feris | Rule-based combination of a hierarchy of classifiers for occlusion detection |
| CN101556717A (en) * | 2009-05-19 | 2009-10-14 | 上海海隆软件股份有限公司 | ATM intelligent security system and monitoring method |
| CN102289672A (en) * | 2011-06-03 | 2011-12-21 | 天津大学 | Infrared gait identification method adopting double-channel feature fusion |
| CN103873825A (en) * | 2014-02-28 | 2014-06-18 | 北京航科威视光电信息技术有限公司 | ATM (automatic teller machine) intelligent monitoring system and method |
| CN104424721A (en) * | 2013-08-22 | 2015-03-18 | 辽宁科大聚龙集团投资有限公司 | Face shield identification method combined with ATM |
-
2016
- 2016-05-31 CN CN201610375049.4A patent/CN106056079B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080247609A1 (en) * | 2007-04-06 | 2008-10-09 | Rogerio Feris | Rule-based combination of a hierarchy of classifiers for occlusion detection |
| CN101556717A (en) * | 2009-05-19 | 2009-10-14 | 上海海隆软件股份有限公司 | ATM intelligent security system and monitoring method |
| CN102289672A (en) * | 2011-06-03 | 2011-12-21 | 天津大学 | Infrared gait identification method adopting double-channel feature fusion |
| CN104424721A (en) * | 2013-08-22 | 2015-03-18 | 辽宁科大聚龙集团投资有限公司 | Face shield identification method combined with ATM |
| CN103873825A (en) * | 2014-02-28 | 2014-06-18 | 北京航科威视光电信息技术有限公司 | ATM (automatic teller machine) intelligent monitoring system and method |
Non-Patent Citations (2)
| Title |
|---|
| DAHUA LIN ET AL.: "Quality-Driven Face Occlusion Detection and Recovery", 《2007 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
| 黄征宇: "用于ATM机遮挡人脸检测的模糊级联分类器和ORB算法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107316312A (en) * | 2017-06-30 | 2017-11-03 | 深圳信路通智能技术有限公司 | A kind of video image occlusion detection method and system |
| CN107633204B (en) * | 2017-08-17 | 2019-01-29 | 平安科技（深圳）有限公司 | Face occlusion detection method, apparatus and storage medium |
| CN107633204A (en) * | 2017-08-17 | 2018-01-26 | 平安科技（深圳）有限公司 | Face occlusion detection method, apparatus and storage medium |
| CN107958572B (en) * | 2017-11-30 | 2019-12-10 | 深圳市百川安防科技有限公司 | Baby monitoring system |
| CN108009496A (en) * | 2017-11-30 | 2018-05-08 | 西安科锐盛创新科技有限公司 | Face blocks recognition methods |
| CN107958572A (en) * | 2017-11-30 | 2018-04-24 | 西安科锐盛创新科技有限公司 | A kind of baby monitoring systems |
| CN107909065A (en) * | 2017-12-29 | 2018-04-13 | 百度在线网络技术（北京）有限公司 | The method and device blocked for detecting face |
| CN107909065B (en) * | 2017-12-29 | 2020-06-16 | 百度在线网络技术（北京）有限公司 | Method and device for detecting face occlusion |
| CN108734927A (en) * | 2018-04-27 | 2018-11-02 | 济南中维世纪科技有限公司 | A kind of electronic peephole viewer blocks alarm system |
| CN108734927B (en) * | 2018-04-27 | 2020-10-02 | 济南中维世纪科技有限公司 | Electronic cat eye shelters from alarm system |
| CN109101923A (en) * | 2018-08-14 | 2018-12-28 | 罗普特（厦门）科技集团有限公司 | A kind of personnel wear the detection method and device of mask situation |
| CN110516649A (en) * | 2019-09-02 | 2019-11-29 | 南京微小宝信息技术有限公司 | Alumnus's authentication method and system based on recognition of face |
| CN110516649B (en) * | 2019-09-02 | 2023-08-22 | 南京微小宝信息技术有限公司 | Face recognition-based alumni authentication method and system |
| CN110647858A (en) * | 2019-09-29 | 2020-01-03 | 上海依图网络科技有限公司 | Video occlusion judgment method and device and computer storage medium |
| CN111951176A (en) * | 2020-06-30 | 2020-11-17 | 重庆灵翎互娱科技有限公司 | Method and equipment for removing local shadows of face image |
| CN112487921A (en) * | 2020-11-25 | 2021-03-12 | 奥比中光科技集团股份有限公司 | Face image preprocessing method and system for living body detection |
| CN112487921B (en) * | 2020-11-25 | 2023-09-08 | 奥比中光科技集团股份有限公司 | Face image preprocessing method and system for living body detection |
| CN112651322A (en) * | 2020-12-22 | 2021-04-13 | 北京眼神智能科技有限公司 | Cheek shielding detection method and device and electronic equipment |
| CN112651322B (en) * | 2020-12-22 | 2024-05-24 | 北京眼神智能科技有限公司 | Cheek shielding detection method and device and electronic equipment |
| CN112801963A (en) * | 2021-01-19 | 2021-05-14 | 湖南优美科技发展有限公司 | Video image occlusion detection method and system |
| CN114708657A (en) * | 2022-03-30 | 2022-07-05 | 深圳可视科技有限公司 | Student attention detection method and system based on multimedia teaching |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106056079B (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106056079B (en) | A kind of occlusion detection method of image capture device and human face five-sense-organ | |
| CN106778595B (en) | Method for detecting abnormal behaviors in crowd based on Gaussian mixture model | |
| CN106006266B (en) | A kind of machine vision method for building up applied to elevator safety monitoring | |
| KR101735365B1 (en) | The robust object tracking method for environment change and detecting an object of interest in images based on learning | |
| CN108062542B (en) | Method for detecting shielded human face | |
| KR102217253B1 (en) | Apparatus and method for analyzing behavior pattern | |
| CN111091098B (en) | Training method of detection model, detection method and related device | |
| CN101556717A (en) | ATM intelligent security system and monitoring method | |
| WO2019114145A1 (en) | Head count detection method and device in surveillance video | |
| CN105303191A (en) | Method and apparatus for counting pedestrians in foresight monitoring scene | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| CN105893946A (en) | Front face image detection method | |
| CN103605971A (en) | Method and device for capturing face images | |
| CN111738342A (en) | Pantograph foreign matter detection method, storage medium and computer equipment | |
| CN108647575A (en) | Drowned method for early warning based on optical visual analysis | |
| CN113362374A (en) | High-altitude parabolic detection method and system based on target tracking network | |
| CN112183472A (en) | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet | |
| Balasundaram et al. | Abnormality identification in video surveillance system using DCT | |
| KR20090044957A (en) | Theft and left baggage survellance system and meothod thereof | |
| Al-Hmouz et al. | License plate localization using a statistical analysis of Discrete Fourier Transform signal | |
| CN110175553B (en) | Method and device for establishing feature library based on gait recognition and face recognition | |
| KR20200060868A (en) | multi-view monitoring system using object-oriented auto-tracking function | |
| Bhaskar et al. | Advanced algorithm for gender prediction with image quality assessment | |
| CN116189063B (en) | Key frame optimization method and device for intelligent video monitoring | |
| CN110502995B (en) | Driver yawning detection method based on fine facial action recognition |
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 |