full term paper of ai

8/7/2019 full term paper of ai

http://slidepdf.com/reader/full/full-term-paper-of-ai 1/14

TERM PAPER

OF

ARTIFICIAL INTELLIGENCE

TOPIC: FACE RECOGNITION

DOS: 20th oct. 2009

SUBMITTED TO:

MISS BALJINDER

SUBMITTED BY:

ARCHANA SINHA

ROLLNO: RH1801B57

REG.NO:10807781

BRANCH-B.TECH(CSE)

SECTION: H1801

GROUP-G2





TOPIC: FACE RECOGNITION

CONTENTS:-----------

INTRODUCTION

WHAT IS FACE RECOGNITION??????

j BIOMETRIC

j 3D FACE RECOGNITION

j Face detection

j Image processing

HISTORY OF FACE RECOGNITION EARLY DEVELOPMENT OF FACE RECOGNITION

STEPS REQUIRED IN FACIAL RECOGNITION

TECHNIQUES USED BY FACE RECOGNITION

TECHNOLOGY TREND OF FACE RECOGNITION

WHY DO HUMAN USE FACE RECOGNITION

ADVANTAGE OF FACIAL RECOGNITION

TECHNICIAL DIFFICULTIESWITH FACE RECOGNITION

HUMAN DIFFICULTIESWITH FACE RECOGNITION

REFERENCE



INTRODUCTION

AI is a branch of computer science concerned with teaching computers to think. It

has had some success in areas like static and moving image recognition but less soin for example playing Chess. It has taken over 40 years to create computer Chess

players able to beat Chess Grandmasters, but that's down to faster CPUs not

software.

In computer games, AI techniques are used to find

shortest paths through complex environments anticipate human player's moves and

fight like a human opponent, adapting to your every move.

A facial recognition system is a computer application

for automatically identifying or verifying a person from a digital image or a videoframe from a video source. One of the ways to do this is by comparing selected

facial features from the image and a facial database. Facial database uses

information regarding to detection of face.

Face recognition is not perfect and struggles to

perform under certain conditions. Ralph Gross, a researcher at the Carnegie Mellon

Robotics Institute, describes one obstacle related to the viewing angle of the face:

"Face recognition has been getting pretty good at full frontal faces and 20 degrees

off, but as soon as you go towards profile, there've been problems."

Early face-detection algorithms focused on the

detection of frontal human faces, whereas newer algorithms attempt to solve themore general and difficult problem of multi-view face detection. That is, the

detection of faces that are either rotated along the axis from the face to the observer

(in-plane rotation), or rotated along the vertical or left-right axis (out-of-plane

rotation), or both.



WHAT IS AI????

AI is a branch of computer science concerned with teaching computers to think. It

has had some success in areas like static and moving image recognition but less so

in for example playing Chess. It has taken over 40 years to create computer Chess

players able to beat Chess Grandmasters, but that's down to faster CPUs notsoftware.

In computer games, AI techniques are used to find shortest paths

through complex environments anticipate human player's moves and fight like ahuman opponent, adapting to your every move.

WHAT IS FACE RECOGNITION??????

A facial recognition system is a computer application for automatically identifying

or verifying a person from a digital image or a video frame from a video source.

One of the ways to do this is by comparing selected facial features from the image

and a facial database. Facial database uses information regarding to detection of

face.

PICTURES REGARDING TO FACE RECOGNITION:-

HISTORY OF FACE RECOGNITION

Face recognition is as old as computer vision, both because of the practical

importance of the topic and theoretical interest from cognitive scientists.

Despite the fact that other methods of can be more accurate, face recognition



has always remains a major focus of research because of its non-invasive

nature and because it is people's primary method of person identification.

Perhaps the most famous early example of a face recognition system is due

to Kohonen who demonstrated that a simple neural net could perform face

recognition for aligned and normalized face images. The type of network he

employed computed a face description by approximating the eigenvectors of

the face image's autocorrelation matrix; these eigenvectors are now known

as `eigenfaces.'

Kirby and Sirovich (1989) later introduced an algebraic manipulation which

made it easy to directly calculate the eigenfaces, and showed that fewer than

100 were required to accurately code carefully aligned and normalized face

images.

Turk and Pentland (1991) then demonstrated that the residual error when

coding using the eigenfaces could be used both to detect faces in cluttered

natural imagery, and to determine the precise location and scale of faces inan image. They then demonstrated that by coupling this method for detecting

and localizing faces with the eigenface recognition method, one could

achieve reliable, real-time recognition of faces in a minimally constrained

environment.

This demonstration that simple, real-time pattern recognition techniques

could be combined to create a useful system sparked an explosion of interest

in the topic of face recognition.

STEPS REQUIRED IN FACIAL RECOGNITION1. Capture image:-First, an image of the face is acquired. This acquisition can

be accomplished by digitally scanning an existing photograph or by using an

electro-optical camera to acquire a live picture of a subject. As video is a rapid

sequence of individual still images, it can also be used as a source of facial

images. 2. Find face in image:-for finding face in image the Second step, software is

employed to detect the location of any faces in the acquired image. This task is

difficult, and often generalized patterns of what a face ³looks like´ (two eyes

and a mouth set in an oval shape) are employed to pick out the faces. 3. Extract features :-Once the facial detection software has targeted a face, it

can be analyzed. As noted in slide three, facial recognition analyzes the spatial

geometry of distinguishing features of the face. Different vendors use different

method to extract the identifying features of a face. Thus, specific details on the

methods are proprietary. The most popular method is called Principle



Components Analysis (PCA), which is commonly referred to as the eigenface

method.

4. Compare templates:-The fourth step is to compare the template generated

in step three with those in a database of Known faces. In an identification

application, this process yields scores that indicate how closely the generatedtemplate matches each of those in the database. In a verification application,

the generated template is only compared with one template in the database ±

that of the claimed identity.

5. Declare matches The final step is determining whether any scores produced

in step four are high enough to declare a match. The rules governing the

declaration of a match are often configurable by the end user, so that he or she

can determine how the facial recognition system should behave based on

security and operational considerations.ind face in image

FIG:-STEPS IN FACE RECOGNITION

BIOMETRIC:-DEFINITION OF BIOMETRIC:-

Any automatically measurable, robust and distinctive physical characteristic or

personal

trait that can be used to identify an individual or verify the claimed identity .

Basically it is used for authentication and verification

Biometrics are used for human recognition which consists of identification

and verification .

Examples of Biometrics IRIS SCAN

RETI NAL SCAN

SPEAKER / VOICE

FI NGERPRI NT

HAND / FI NGER GEOMETRY



SIG NATURE VERIFICATIO N

KEYSTROKE DY NAMICS

OTHER ESOTERIC BIOMETRICS

GAIT

EAR ODOR

IMAGE PROCESSINGIMAGE PROCESSI NG is any form of signal processing for which the input is

animage, such as photographs or frames of video; the output of image processing

can be either an image or a set of characteristics or parameters related to the image.

IMAGE PROCESSING OPERATIONS ARE:

Euclidean geometry transformations such as enlargement, reduction,

and rotation

Color corrections such as brightness and contrast adjustments, quantization,

or color translation to a different color space

Digital compositing or optical compositing (combination of two or more

images). Used in film-making to make a "matte"

Interpolation, demosaicing, and recovery of a full image from a raw image

format using a Bayer filter pattern

Image registration, the alignment of two or more images

Image differencing and morphing Image recognition, for example, extract the text from the image by using optical

character recognition

Image segmentation



High dynamic range imaging by combining multiple images

Geometric hashing for 2-D object recognition with affine invariance

Three-dimensional face recognition :-

j (3D face recognition) is a modality of facial recognition methods in which

the three-dimensional geometry of the human face is used.

j 3D face recognition methods can achieve significantly higher accuracy than

their 2D counterparts, rivaling fingerprint recognition.

j 3D face recognition has the potential to achieve better accuracy than its 2D

counterpart by measuring geometry of rigid features on the face. This avoids

such pitfalls of 2D face recognitionalgorithms as change in lighting,

different facial expressions, make-up and head orientation.

j Another approach is to use the 3D model to improve accuracy of traditional

image based recognition by transforming the head into a known view.



j most range scanners acquire both a 3D mesh and the corresponding texture.

This allows combining the output of pure 3D matchers with the more

traditional 2D face recognition algorithms, thus yielding better performance

j The main technological limitation of 3D face recognition methods is the

acquisition of 3D images, which usually requires a range camera. This isalso a reason why 3D face recognition methods have emerged significantly

later (in the late 1980s) than 2D methods. Recently commercial solutions

have implemented depth perception by projecting a grid onto the face and

integrating video capture of it into a high resolution 3D model.

j This allows for good recognition accuracy with low cost off-the-

shelf components.

FACE DETECTION

j Face detection is a computer technology that determines the locations andsizes of human faces in arbitrary (digital) images. It detects facial features

and ignores anything else, such as buildings, trees and bodies.

j Face detection can be regarded as a specific case of object-class detection; In

object-class detection, the task is to find the locations and sizes of all objects

in an image that belong to a given class. Examples include upper torsos,

pedestrians, and cars.

j In face detection, one does not have this additional information.

j Early face-detection algorithms focused on the detection of frontal humanfaces, whereas newer algorithms attempt to solve the more general and

difficult problem of multi-view face detection. That is, the detection of faces

that are either rotated along the axis from the face to the observer (in-plane

rotation), or rotated along the vertical or left-right axis (out-of-plane

rotation), or both.

. TECHNIQUES USED BY FACE RECOGNITION

Face recognition uses mainly the following techniques: ----------------- Facial geometry: uses geometrical characteristics of the face. May use

several cameras to get better accuracy (2D, 3D...)

Skin pattern recognition (Visual Skin Print)

Facial thermogram: uses an infrared camera to map the face temperatures

Smile: recognition of the wrinkle changes when smiling

Compare tem



TECHNOLOGY TREND OF FACE RECOGNITION

EIGENFACE METHOD:- This technology re-constructs a face by

superimposing a set

of so-called ³eigenface´. The similarity of two facial images is determined based

on the coefficient of the relevant eigenfaces

Local Feature Analysis This technology locally uses the eigenface method for

a few

single parts of the face (e.g. eyes, nose, mouth) and additionally determines their

geometric proportions to each other.

Dynamic Local Feature Analysis (³DLFA´):- Based on the neural

network-based classification algorithm, the face is represented by an ³dynamic´

edge analyzing the facial shape and texture and thus basing the comparison on up

to 108 characteristics.

Why do human use face recognition? Universal. Everyone has a face, everyone can enrol.

Non-intrusive. A non-contact verification process, similar to having a photo

taken.

Incredibly Fast. Responsive software ³finds´ the user¶s face within a frame

and starts making comparisons instantly. Sensors don¶t need to be adjusted for

height; spectacles and safety gloves don¶t need to be removed; operatives don¶tneed to touch a scanner or present their face in a particular way. Verification

takes approximately 1.5 seconds.

Accurate. Better than humans at verifying identity, and able to work 24 hours a

day.

Dependable. Successfully deployed in challenging real-life environments,

overcoming usual biometric obstacles such as dust, dirt, grease, variable lighting

and user co-operation. Aurora systems are in use in over 940 construction

locations in the UK and Middle East, processing over fifteen million clocking

actions a year.

Transparent. Face recognition is the only biometric where the transaction

records can be visually confirmed.

ADVANTAGE OF FACIAL RECOGNITION pj It is mainly used in airports were it will recognize the face and we can avoid

some unwanted terrorist.



j Uses faces, which are public

j Involves non-intrusive, contact-free process

j Uses legacy databases

j Integrates with existing surveillance systems

TECHNICIAL DIFFICULTIES WITH FACE RECOGNITION

1. Finding Faces

y Uncontrolled background

y Subject¶s non-cooperation

Subject not looking at camera

Subject wearing hat, sunglasses, etc.

y Moving target

2. Identifying Faces

y Uncontrolled environmental conditions

Lighting (shadows, glare)

Camera angle

Image resolution

Human Difficulties with Facial Recognition

1. Inherent Operator Limitations

y Humans are not good at recognizing faces of

people they do not know

2. Operator Overload Vast amounts of information

Limited attention span

Limited accuracy

3. Operator Reliability

y Dedication

y Honesty

WEAKNESSES OF FACE RECOGNITION:-

j Face recognition is not perfect and struggles to perform under certain

conditions. Ralph Gross, a researcher at the Carnegie Mellon Robotics

Institute, describes one obstacle related to the viewing angle of the face:

"Face recognition has been getting pretty good at full frontal faces and 20

degrees off, but as soon as you go towards profile, there've been problems."



j face recognition does not work well include poor lighting, sunglasses, long

hair, or other objects partially covering the subject¶s face, and low resolution

images.

j Another serious disadvantage is that many systems are less effective if facial

expressions vary. Even a big smile can render in the system less effectively.For instance: Canada now allows only neutral facial expressions in passport

photos.



REFERENCE:-

1. WWW.GOOGLE.COM 2. WWW.WIKIPEDIA.COM 3. WWW.AMEZAON.COM

full term paper of ai

Documents