Sie sind auf Seite 1von 16

Base paper Titles: Ocular Recognition for Blinking Eyes

Modified Title: Visual Image Recognition for Intermittent Eyes

Abstract

Nowadays, people spend more time in front of electronic screens like


computers, laptops, TV screens, mobile phones or tablets which cause eye blink
frequency to decrease. Each blink spreads the tears on the eye cornea to moisture and
disinfects the eye. Reduced blink rate causes eye redness and dryness also known as
Dry Eye, which belongs to the major symptoms of the Computer Vision Condition.
The goal of this work is to design eye blink detector which can be used in dry eye
prevention system. We have analyzed available techniques for blink detection and
designed our own solutions based on histogram back projection and optical flow
methods. We have tested our algorithms on different datasets under various lighting
conditions. Inner movement detection method based on optical flow performs better
than the histogram based ones. Finally, the non-linear SVM is applied for
identification purpose. Experimental results demonstrate that the proposed algorithm
achieves the best accuracy for both open eye and blinking eye scenarios. It achieve
higher recognition rate and much lower false positive rate than the-state-of-the-art
technique presented it offers greater flexibility for the prospective subjects during
recognition as well as higher reliability for security.

Aim and Objective

Our aim is to create eye blink detector, which could be used in real-time blink
detection system. In case of low blink rate it will notify a user to blink more
frequently. This paper proposes two different methods on blink detection. The first of
presented algorithms computes back projection from 1D saturation and 2D hue-
saturation histogram. The second method addressed as Inner movement detection
detects eyelid motion using Lucas-Kanade (KLT) feature tracker
Existing System

In recent years, biometric has become an important part of various


applications, including identification, video surveillance, and recognition system.
The reason behind is that it exploits the bio-invariant characteristics for yielding
robust features such as the eye region which is explored in the iris and ocular
recognitions. Comparing with the aforementioned two biometrics, benefits of ocular
recognition are as follows: 1) Iris recognition has a relatively high discriminative
capability. 2) Eye region provides eye shape information which is useful for
identification. 3) Ocular recognition has a greater tolerance in handling a broad range
of distances, and; 4) Eyes recognition can still cope with partially covered faces on
particular regions, such as nose and mouth.

Drawback in Existing System

Proposed System

It focuses on ocular recognition to compensate the abovementioned issue and


the shortcomings of iris recognition. In the proposed algorithm, the landmark points
of eyes are firstly detected. To suppress the influences of skin colors and lighting
conditions, the geometric feature, which particularly describes the contours of eyes,
are considered. The weighted texture-based LBP (WT-LBP) is proposed to extract
the texture property of the selected regions. In addition, the probabilities of single-
and double-fold eyelids are also derived for description. To further enhance the
performance, the combination of various features with the sequential forward floating
selection (SFFS) is further utilized. Finally, the non-linear support vector machine
(SVM) is applied for classification purpose. Experimental results suggest that the
proposed ocular recognition method achieves excellent performance on three
different face databases, which in turn suggests that the proposed method is an
attractive candidate for practical biometrics applications. Possible improvements
include the exploration of more robust features to further enhance the performance on
the blinking and closed eye scenarios.
Advantages

 The triggering event can be set as the user wants it. For instance – If a user
wants to trigger image capture when he blinks his left eye twice, he should be
able to set it.
 This device can be used as spying equipment, which can help spies,
detectives and officials capture certain images secretly without losing focus
of their target.
 Cops can use this device to scan people on the go and prevent harmful attacks
by people carrying harmful weapons illegally. For this purpose, the glasses in
the device can be converted to a LCD screen and Image Processing
techniques can be used to detect any substance they want to.
 It can also be used for video recording purposes by students, professors and
many other sections of the society.
Literature survey

Title: Impact of The Lips For Biometrics


Author: y. F. Liu, c. Y. Lin and j. M. Guo
Year: 2013

 In fact, it is a challenging issue for identity recognition solely by the lips.


 In the first stage of the proposed system, a fast box filtering is proposed to
generate a noise-free source with high processing efficiency.
 The feature extraction, two geometric ratios and ten parabolic-related
parameters are adopted for further recognition through the support vector
machine.
 Experimental results demonstrate that, when the number of subjects is fewer
or equal to 29, the correct accept rate (car) is greater than 98%, and the false
accept rate (far) is smaller than 0.066%. Moreover, the processing speed of
the overall system achieves 34.43 frames per second, which meets the real-
time requirement.

Advantage:
 Lips biometrics is passive biometrics - user/subject interaction is not needed.
Images may be acquired from the distance without the knowledge of the
examined person.
 Lips biometrics is anatomical - better results are expected than in behavioral
biometrics. Lips are usually visible - not hidden/overcast by anything.
 Lips may be implemented in hybrid lips-face or lips-voice biometric systems.

Disadvantage:
 Using such biometric identifier, the 22 devices used to measure the biometric
 The face needs to be well lighted by controlled light sources in the first place
in order to acquire high quality images
Title: Face Recognition By Humans: Nineteen Results All Computer Vision
Author: p. Sinha, b. Balas, y. Ostrovsky and r. Russel
Year: 2006.

 These findings provide insights into the nature of cues that the human visual
system relies upon for achieving its impressive performance and serve as the
building blocks for efforts to artificially emulate these abilities.
 It present what we believe are 19 basic results, with implications for the
design of computational systems. Each result is described briefly and
appropriate pointers are provided to permit an in-depth study of any particular
result.
Advantage:
 Human-Computer Interaction
 It can be used as a part of identification and access control systems in
organizations, as well as personal devices, such as in the case of smartphones.
Disadvantage
 Issues About Reliability and Efficiency
 Its Biometric Information Privacy Act requires affirmative consent for
companies to collect biometric data.
Title: On The Fusion of Periocular and Iris Biometrics In Non-Ideal Imagery
Author: d. L. Woodard, s. Pundlik and p. Miller
Year: 2010
 The periocular region surrounding the iris, along with the iris texture patterns,
in order to improve the overall recognition performance in such images.
 Periocular texture is extracted from a small, fixed region of the skin
surrounding the eye.
 The Experiments on the images extracted from the near infrared (nir) face
videos of the multi biometric grand challenge (mbgc) dataset demonstrate that
valuable information is contained in the periocular region and it can be fused
with the iris texture to improve the overall identification accuracy in non-
ideal situations.
Advantages:
 The use of shape-based features extracted from the periocular region for
biometric recognition.
 eyebrow shape features may be more robust to varying illumination and
image blur due to motion

Disadvantage:
 The use of SIFT features is the amount of time required for computation
 Less-constrained biometric recognition scenarios, the image quality may not
be adequate to extract a significant number of key-point-based feature.
Title: A Colour Iris Recognition System Employing Multiple Classifier Techniques
Author: p. Radu, k.sirlantiz, g. Howells, s. Hoque and f. Deravi
Year: 2013
 In the present work we propose an iris recognition methodology which is
designed to cope with noisy colour iris images.
 Two main contributions of this paper: first, we adapt standard iris features
proposed in literature for near infrared images by applying a feature selection
method on features extracted from various color channels.
 Second, multiple classifier system architecture to enhance the recognition
accuracy of the biometric system. With a feature size of only 360 real valued
components, the proposed iris recognition system performs with a high
accuracy on ubirisv1 dataset, in both identification and verification scenarios.

Advantage:
 Highly protected, internal organ of the eye
 Encoding and decision-making are tractable
 Externally visible; patterns imaged from a distance
Disadvantage

 Located behind a curved, wet, reflecting surface


 Partially occluded by eyelids, often drooping
 Illumination should not be visible or bright
Title: Pigment Melanin: Pattern For Iris Recognition
Author: m. S. Hosseini, b. N. Araabi, and h. Soltanian-zadeh
Year: 2010
 To describe the patterns, a shape analysis method is used to derive the feature
code for each subject. An important question is how the melanin patterns,
which are extracted from vl, are independent of the iris texture in nir. With
this question in mind, the present investigation proposes fusion of features
extracted from nir and vl to boost recognition performance.
 We have collected our own database (utiris), consisting of both nir and vl
images of 158 eyes of 79 individuals. This investigation demonstrates that the
proposed algorithm is highly sensitive to the patterns of chromophores and
improves the iris recognition rate.

Advantage

 Accuracy
 highly scalable and can be used in both large and small scale programs
 Iris patterns remain stable throughout an individual’s life. It is protected by
the body’s own mechanism

Disadvantage
 A person has to be steady in front of the device to be enrolled by iris
scanners.
 it is hard to perform an iris scanning due to the presence of reflections
Title: Mitigating Effects of Plastic Surgery: Fusing Face And Ocular Biometrics
Author: r. Jillela and a. Ross
Year: 2012.

 The task of successfully matching face images obtained before and after
plastic surgery is a challenging problem.
 The degree to which a face is altered depends on the type and number of
plastic surgeries performed, and it is difficult to model such variations.
 The proposed approach provides the highest reported recognition
performance on a publicly accessible plastic surgery database, with a rank-
one accuracy of 87.4%.

Advantage

 The iris is an internal organ of the eye. For this reason, iris pattern never
changes in a lifetime.
 The encoding and decision making of iris pattern is highly traceable. It takes
only 30 milliseconds for the image analysis and the subsequent encoding.

Disadvantage

 It is obscured by eyelashes, lenses and reflections, which create a problem,


more often than not.
 Iris is partially blocked by eyelids which is difficult to control by individuals
due to frequent blinking.
SOFTWARE SPECIFICATION

Software Requirements:

 Windows 7 or higher
 Matlab
 Simulink

Hardware Components:

 Processor – i3
 Hard Disk – 160 GB
 Memory – 2GB RAM
DETAILED DESIGN

System Architecture

Training Phase

Test Phase
Data Flow Diagram
Modules Description

Image Features

In order to classify a part of a photo it is necessary to get certain features of it.


They are the base for object classification. The most well-known and important
visual feature extractions are

Histogram Backprojection

it use histogram to represent skin color of the user. Histogram backprojection


creates a probability map over the image. In other words backprojection determines
how well the pixels from the image fit the distribution of a given histogram. Higher
value in a backprojected image denotes more likely location of the given object. We
detect closed eyes as high percentage of skin color pixels within the eye region
otherwise we consider eyes opened

The HSV (Hue Saturation Value) color model to achieve partial luminance
invariance by the omission of the Value channel. We did experiments with two
different histograms:

 1D saturation histogram (histogram S),


 2D hue-saturation histogram (histogram HS)

First we detect the user’s face by Haar Cascade Classifier. We calculate the skin
color histogram from a sequence of images of face regions. Other parts of the image
are not used to obtain as precise skin color histogram as possible. Histogram is
normalized afterwards and regularly updated. For every input image we calculate the
backprojection with this histogram. Subsequently a resulting backprojected image is
modified using morphological operations (Open and Erode) and threshold (threshold
HS = 10 in hue-saturation and threshold S = 25 in saturation histogram obtained by
experiments) to increase small difference between open and closed eyelids due to
lower skin probability of eyelids caused by shadows in eye areas or make-up. Finally
the average value of the probabilities is calculated from the region of the user’s eyes.
Pixel-Based Movement Detection

Initially, I used the algorithm to detect a face and later to detect eyes. The
main idea behind that was to take the first frame as a background. In the first frame
eyes must be open. Every next frame is compared with the first one and the algorithm
checks how many pixels have been changed

Inner Movement Detection

Our own Inner Movement Detection algorithm based on optical flow. Optical
flow locates new feature position in the following frame. One of the most common
methods called KLT tracker selects features suitable for tracking with high intensity
changes in both directions. If a user blinks, the mean displacement of feature points
within the eye region should be greater than the displacement of the rest of the points
within the face area. The first step consists of localizing a user’s face and eyes using
Haar Cascade Classifier on grayscale image. We initialize random KLT features
within the eye and nose regions and classify them as left ocular, right ocular or non-
ocular. These features are being tracked by KLT tracker. Tracking is reinitialized in
regular intervals or in case of loss of many feature points. We compute the average
displacement separately for three groups of points. Afterwards we compare the
difference between the left or right ocular and non-ocular movement displacements.
If this difference exceeds a threshold value (threshold diff = face. height/165, where f
ace. height is the height of detected face in the initial phase), a movement within the
eye region is anticipated.

Eye-blink detection system

Vision-based eye-blink monitoring systems have many possible applications,


like fatigue monitoring, human–computer interfacing and lie detection. No matter
what the purpose of the system is, the developed algorithm must be reliable, stable
and work in real time in varying lighting conditions.
Prediction during Tracking Phase

our eye detection algorithm during the Tracking Phase. Due to the movement of
user’s hand that holds the mobile device, the eye frequently falls out of the Search
Region. In this case, existing schemes discard the current video frame and try to find
eyes in the Detection Phase, resulting in severe delays. If the eye is not found in the
Search Region during the Tracking Phase, our algorithm tries to find the eye in the
current frame once more based on the information of vertical and horizontal tilt angle
difference

 The vertical tilt angle of the device remains between 0°and 90°in most cases.
 The user maintains a certain distance to the device, thus the size of one eye
does not exceed 80% of the captured camera frame
 Depending on the position of the camera, the location of the eyes changes.

Das könnte Ihnen auch gefallen