CN106682588A - Real-time pupil detection and tracking method - Google Patents
Real-time pupil detection and tracking method Download PDFInfo
- Publication number
- CN106682588A CN106682588A CN201611108805.3A CN201611108805A CN106682588A CN 106682588 A CN106682588 A CN 106682588A CN 201611108805 A CN201611108805 A CN 201611108805A CN 106682588 A CN106682588 A CN 106682588A
- Authority
- CN
- China
- Prior art keywords
- real
- image
- face
- tracking
- time
- 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
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
- G06V40/18—Eye characteristics, e.g. of the iris
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Ophthalmology & Optometry (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a real-time pupil detection and tracking method. The method includes: (1) adopting a camera to shoot facial images in real time; (2) adopting an image capture card for capturing video images to obtain a sequence of a group of images; (3) subjecting each frame of captured images to skin detection to determine whether a human face exists or not and judge an approximate area of the human face; (4) detecting a pupil position in the area of the human face and tracking. By a YCbCr color space based skin color model, the method is high in detection rate under conditions of different skin colors and large illumination changes.
Description
Technical field
The application is related to vehicle security drive technical field, more particularly to a kind of real-time pupil detection and tracking.
Background technology
Pupil detection is suggested as a part for recognition of face.In terms of pupil detection, although having many both at home and abroad
People proposes different viewpoints, and the method for current main flow is had template matching method, sciagraphy, " eigen eyes " method, become based on small echo
Change method, symmetry method and the method based on neutral net.
Kanade and other researchers develop the method that pupil feature is detected from a width still image, more pupils
Hole characteristic detection method is the irregular template method based on YUlll, but because the scheme of irregular template needs to expend big
The time of amount, the pupil tracking in video image is difficult to apply in this way.But if tracked for pupil tracking, this
A little schemes are also far from meeting requirement of real-time.
With each research and raising of the application field to requirement of real-time, to real-time pupil detection and tracking system
The requirement of research of uniting also increasingly looms large and urgently.But so far can also be accurate without a kind of perfect theoretical and method
Really carry out real-time, accurate detection and track come the pupil to face in video image with effective method.
Accurately to detect and can be in real-time tracking to image pupil, it is necessary to determine the face area in image first
Domain, the method for then further being detected in human face region again, positioning pupil, so can not only improve the precision of detection, more
The region of pupil detection can be substantially reduced, detection speed is improved.
Face datection problem is one the problem of larger challenge, because face includes the irregular such as face, hair
Target complicated to be measured, and the image of collection is highly susceptible to the change of various conditions and various influence of noises, main performance
For:
(1), influenceed by factors such as appearance, expression, the colours of skin, face has very big changeability;
(2) adjuncts such as glasses, beard and ornament, be there may be on the face in many cases;
(3), head movement can cause face location, angle etc. in image to change;
(4), different illumination conditions produce influence to facial image in image, and brightness excursion is very big.
(5), facial size, position and ambient noise etc. are different in different images.
The content of the invention
It is an object of the invention to provide a kind of real-time pupil detection and tracking, with overcome it is of the prior art not
Foot.
To achieve the above object, the present invention provides following technical scheme:
The embodiment of the present application discloses a kind of real-time pupil detection and tracking, including:
(1), camera shoots facial image in real time;
(2) video image, is gathered with image pick-up card, one group of image sequence is obtained;
(3), each two field picture for collecting is carried out skin detection to determine whether face and judge face first
Approximate region;
(4) pupil position, is detected in human face region and is tracked.
Preferably, in above-mentioned real-time pupil detection and tracking, the step (3) includes:
S1, face database is set up, the database includes the facial image under different light conditions,
The distribution situation of face skin color chromatic component in YCbCr color spaces in s2, analytical database, and set up
Chromatic component is distributed graph model;
S3, the coloured image to detecting carry out shape filtering;
S4, filtered coloured image is converted into bianry image.
Preferably, in above-mentioned real-time pupil detection and tracking, in the step s1, picture format is 24 positions
Figure.
Preferably, in above-mentioned real-time pupil detection and tracking, the step (1) includes:
Infrared light supply produces two kinds of infrared lights of wave band of 850nm and 900nm;
Two kinds of wavelength servant's face images of 850nm and 900nm are gathered respectively using two cameras.
Compared with prior art, the advantage of the invention is that:Complexion model of the present invention based on YCbCr color spaces.This
Inventive method is respectively provided with preferable verification and measurement ratio under the different colours of skin and larger Varying Illumination.
Brief description of the drawings
In order to illustrate more clearly of the embodiment of the present application or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments described in application, for those of ordinary skill in the art, on the premise of not paying creative work,
Other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 show the flow chart of real-time pupil detection and tracking in the specific embodiment of the invention;
Fig. 2 show the functional-block diagram of real-time pupil tracing system in the specific embodiment of the invention.
Specific embodiment
Eye Tracking Technique by the use of some constant eye structures of relative position during Rotation of eyeball and feature as reference,
Sight line running parameter is extracted between change in location feature and these invariant features, is then obtained by geometrical model or mapping model
Direction of visual lines, by infrared illumination, produces the infrared light of different-waveband, reflectivity of the eye retina to different wave length infrared light
Low, the low eye image of reflectivity is dark, and reflectivity eye image high is bright.
Below in conjunction with the accompanying drawing in the embodiment of the present invention, detailed retouching is carried out to the technical scheme in the embodiment of the present invention
State, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on the present invention
In embodiment, the every other implementation that those of ordinary skill in the art are obtained on the premise of creative work is not made
Example, belongs to the scope of protection of the invention.
With reference to shown in Fig. 1, real-time pupil detection and tracking, including:
(1), camera shoots facial image in real time;
(2) video image, is gathered with image pick-up card, one group of image sequence is obtained;
(3), each two field picture for collecting is carried out skin detection to determine whether face and judge face first
Approximate region;
(4) pupil position, is detected in human face region and is tracked.
Further, step (3) includes:
S1, face database is set up, the database includes the facial image under different light conditions,
The distribution situation of face skin color chromatic component in YCbCr color spaces in s2, analytical database, and set up
Chromatic component is distributed graph model;
S3, the coloured image to detecting carry out shape filtering;
S4, filtered coloured image is converted into bianry image.
Bianry image after completion skin detection can deposit error in both cases, and a kind of is the introducing quilt due to noise
It is mistaken for the isolated point or fritter isolated area of the colour of skin;Another kind is because the colour of skin point in the human face region in image is existed
Certain difference, is mistaken for non-colour of skin point.Form must be carried out to eliminate these errors to carrying out the image after skin detection
Filtering.
The mistake that can be eliminated during some skin detections by shape filtering is judged to the non-skin point of skin points.
Preferably, in step s1, picture format is 24 bitmaps.
Further, step (1) includes:
Infrared light supply produces two kinds of infrared lights of wave band of 850nm and 900nm;
Two kinds of wavelength servant's face images of 850nm and 900nm are gathered respectively using two cameras.
With reference to shown in Fig. 2, real-time pupil tracing system, including:
First infrared light supply, produces the infrared light of 900nm wave bands;
Second infrared light supply, produces the infrared light of 850nm wave bands;
First video acquisition module, the image of the first infrared light supply reflection on collection face;
Second video acquisition module, the image of the second infrared light supply reflection on collection face;
Video processing module, differential comparison is carried out to the image that the first infrared light supply and the second infrared light supply reflect, and is positioned
Position of human eye.
In certain embodiments, the first video acquisition module and the second video acquisition module are respectively adopted non-refrigeration focal surface
Thermal imagery component is equipped with infrared lens.
In certain embodiments, video processing module includes FPGA module and SRAM module.
In certain embodiments, also include:
First Video decoding module, between the first video acquisition module and video processing module;
Second Video decoding module, between the second video acquisition module and video processing module.
In certain embodiments, also including video display module, and it is sequentially arranged in video processing module and video shows
Configuration module, video encoding module and data memory module between module.
Application examples 1
Driver's fatigue degree is confirmed using above-mentioned real-time pupil tracing device.Should by the real-time pupil tracing device
In for onboard system, driver's fatigue degree is analyzed, and makes early warning.
Specifically the flow of detection alarm is:Waking state preset parameter->Human eye positioning->Degree of fatigue confirmation->Fatigue is driven
Sail alarm.First, for different people, the analysis of record picture, determines the indices of human eye under waking state in advance, including pupil is big
Small, eye-closing period, open and close eyes ratio;Then the human eye parameter of the picture of analysis record in real time, diminishes when pupil, closed-eye time is long, open and close
The degree of eye ratio reduction, is confirmed whether in fatigue state.
The course of work:
(1) waking state human eye parameter is configured.Under waking state, advance with video acquisition module and gather under two kinds of wavelength
Facial image, Video decoding module is decoded, and obtains waking state indices by graphical analysis, and these data are deposited
Storage is in data memory module;
(2) degree of fatigue confirms.Two kinds of wavelength servant's face images are gathered using video acquisition module, Video decoding module enters
Row decoding, present status indices are obtained by graphical analysis, are made comparisons with preset waking state indices, according to super
The degree for going out confirms to be in the degree of fatigue state, so as to make corresponding alarm, and realizes that module is entered by video
Row display.
Application examples 2
Driving safety is confirmed using above-mentioned real-time pupil tracing device.Above-mentioned real-time pupil tracing device is applied to car
In loading system, driving safety is confirmed by confirming pilot's line of vision lime light, when not according to normal operating, do responding
Report.
Specifically the flow of detection alarm is:
Sight line note preset parameter during all parts (left and right reflective mirror, rearview mirror etc.)->Needed when various driving situations are set
The action of the sight line to be made->Human eye positioning->Driving situation confirmation->Dangerous driving alarm.
First, before driving, parameter is recorded in advance when pilot's line of vision is noted into all parts (left and right reflective mirror, rearview mirror etc.)
Analysis is drawn, then the sight line action that driver during various driving situations needs to make is set according to these basic parameters.It is actual
During driving, determine that the sight line that whether driver makes as required during various driving situations is acted by positioning human eye in real time.
If do not make needing action, alarm.
The real-time pupil tracing device of this case, not only goes for onboard system, has characteristic kinematic rule for other
Or can also be applicable by the equipment that motor pattern extrapolates its motion feature.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality
Body or operation make a distinction with another entity or operation, and not necessarily require or imply these entities or deposited between operating
In any this actual relation or order.And, term " including ", "comprising" or its any other variant be intended to
Nonexcludability is included, so that process, method, article or equipment including a series of key elements not only will including those
Element, but also other key elements including being not expressly set out, or also include being this process, method, article or equipment
Intrinsic key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that
Also there is other identical element in process, method, article or equipment including the key element.
The above is only the specific embodiment of the application, it is noted that for the ordinary skill people of the art
For member, on the premise of the application principle is not departed from, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as the protection domain of the application.
Claims (4)
1. a kind of real-time pupil detection and tracking, it is characterised in that including:
(1), camera shoots facial image in real time;
(2) video image, is gathered with image pick-up card, one group of image sequence is obtained;
(3), carry out skin detection first to each two field picture for collecting to determine whether face and judge the big of face
Cause region;
(4) pupil position, is detected in human face region and is tracked.
2. real-time pupil detection according to claim 1 and tracking, it is characterised in that the step (3) includes:
S1, face database is set up, the database includes the facial image under different light conditions,
The distribution situation of face skin color chromatic component in YCbCr color spaces in s2, analytical database, and set up colourity
Component profile model;
S3, the coloured image to detecting carry out shape filtering;
S4, filtered coloured image is converted into bianry image.
3. real-time pupil detection according to claim 1 and tracking, it is characterised in that:In the step s1, image
Form is 24 bitmaps.
4. real-time pupil detection according to claim 1 and tracking, it is characterised in that:The step (1) includes:
Infrared light supply produces two kinds of infrared lights of wave band of 850nm and 900nm;
Two kinds of wavelength servant's face images of 850nm and 900nm are gathered respectively using two cameras.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611108805.3A CN106682588A (en) | 2016-12-06 | 2016-12-06 | Real-time pupil detection and tracking method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611108805.3A CN106682588A (en) | 2016-12-06 | 2016-12-06 | Real-time pupil detection and tracking method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106682588A true CN106682588A (en) | 2017-05-17 |
Family
ID=58866323
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611108805.3A Pending CN106682588A (en) | 2016-12-06 | 2016-12-06 | Real-time pupil detection and tracking method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106682588A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108960153A (en) * | 2018-07-06 | 2018-12-07 | 深圳虹识技术有限公司 | A kind of method and apparatus of adaptive iris recognition |
CN112633080A (en) * | 2020-12-03 | 2021-04-09 | 昆明依利科特科技有限公司 | Automatic detection system and detection method for passenger virus exposure in vehicle for highway access |
CN112633080B (en) * | 2020-12-03 | 2024-06-07 | 昆明依利科特科技有限公司 | Automatic detection system and detection method for vehicle occupant toxicity in highway bayonet |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810472A (en) * | 2013-11-29 | 2014-05-21 | 南京大学 | Method for pupil position filtering based on movement correlation |
-
2016
- 2016-12-06 CN CN201611108805.3A patent/CN106682588A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810472A (en) * | 2013-11-29 | 2014-05-21 | 南京大学 | Method for pupil position filtering based on movement correlation |
Non-Patent Citations (2)
Title |
---|
党治: "实时瞳孔检测与跟踪方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
刘洪榛: "基于机器视觉的疲劳驾驶检测算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108960153A (en) * | 2018-07-06 | 2018-12-07 | 深圳虹识技术有限公司 | A kind of method and apparatus of adaptive iris recognition |
CN112633080A (en) * | 2020-12-03 | 2021-04-09 | 昆明依利科特科技有限公司 | Automatic detection system and detection method for passenger virus exposure in vehicle for highway access |
CN112633080B (en) * | 2020-12-03 | 2024-06-07 | 昆明依利科特科技有限公司 | Automatic detection system and detection method for vehicle occupant toxicity in highway bayonet |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109643366B (en) | Method and system for monitoring the condition of a vehicle driver | |
Eriksson et al. | Eye-tracking for detection of driver fatigue | |
CN102622588B (en) | Dual-certification face anti-counterfeit method and device | |
JP4811259B2 (en) | Gaze direction estimation apparatus and gaze direction estimation method | |
CN105354987B (en) | Vehicle-mounted type fatigue driving detection and identification authentication system and its detection method | |
EP1732028B1 (en) | System and method for detecting an eye | |
US6920236B2 (en) | Dual band biometric identification system | |
CN104751600B (en) | Anti-fatigue-driving safety means and its application method based on iris recognition | |
US7043056B2 (en) | Facial image processing system | |
CN103714659B (en) | Fatigue driving identification system based on double-spectrum fusion | |
Batista | A drowsiness and point of attention monitoring system for driver vigilance | |
CN105769120A (en) | Fatigue driving detection method and device | |
CN105739705A (en) | Human-eye control method and apparatus for vehicle-mounted system | |
CN105286802B (en) | Driver Fatigue Detection based on video information | |
CN104224204A (en) | Driver fatigue detection system on basis of infrared detection technology | |
CN101593352A (en) | Driving safety monitoring system based on face orientation and visual focus | |
JP2003015816A (en) | Face/visual line recognizing device using stereo camera | |
CN108182377A (en) | Human eye sight detection method and device based on photogrammetric technology | |
CN105844227A (en) | Driver identity authentication method for school bus safety | |
CN106548600A (en) | Real-time pupil tracing system and fatigue state monitoring method | |
CN110363070A (en) | The method, apparatus and computer program product of intelligent recognition road condition | |
CN114821695A (en) | Material spectrometry | |
CN114821696A (en) | Material spectrometry | |
CN113920591A (en) | Medium-distance and long-distance identity authentication method and device based on multi-mode biological feature recognition | |
CN106682588A (en) | Real-time pupil detection and tracking method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170517 |