SASURIE COLLEGE OF ENGINEERING VIJAYAMANGALAM TIRUPUR

URIE

Presenting a paper ADVANCED FACE RECOGNITION TECHNOLOGY

Submitted by
K.U.SUKHANYAA [email protected] S.SUMITHRA sumithrasel [email protected] B.E CSE IIndYEAR

and

neural

networks,

will

also

be

explained.

Abstract
Recently face recognition is

attracting much attention in the society of network multimedia information access. Areas such as network security, content indexing and retrieval, benefits and from video face compression

Definition of Face Recognition

%he ability to recogni'e people by their facial characteristics. advanced technology of a person s face into a multidimensional face space. *omputers can conduct facial database searches and#or perform live, one!to!one or one!to!many verifications with unprecedented accuracy and split! second processing. +sers can be granted secure access to their computer, mobile devices, or for online e!commerce, simply by looking into their ,eb camera. is based on the )igenface algorithm, which maps the characteristics %he most

recognition technology because "people" are the center of attention in a lot of video. Network access control via face recognition not only makes hackers virtually impossible in to steal one s "password", but also increases the user! friendliness human!computer interaction. "ndexing and#or retrieving video data based on the appearances of particular persons will be use!ful for users such as news reporters, political scientists, and moviegoers. $or the applications of videophone and teleconferencing, the assistance "n of this face recognition we of give this also an new provides a more efficient coding scheme. paper, course introductory

information processing technology. %he paper shows the readers the generic framework system, fre&uently recogni'er. for the the face variants by recognition that the are face face and

encountered (everal

famous

recognition algorithms, such as eigenfaces

 3oice!scan (ignature!scan 4eystroke!scan

A 5biometric system6 refers to the integrated hardware and software used to conduct biometric identification or verification.

Why we choose face recognition over other biometric? What are biometrics?
A biometric is a uni&ue, measurable characteristic of a human being that can be used to automatically recogni'e an individual or verify an individual-s measure identity. both .iometrics can and physiological the following 7. "t re&uires no physical inetraction on behalf of the user. 8. "t is accurate and allows for high enrolment and verification rates. 9. "t does not re&uire an expert to interpret the comparison result. :. "t can use your existing hardware infrastructure, existing camaras and image capture devices will work with no problems. %here are a number reasons to choose face recognition. %his includes

behavioral characteristics. Physiological biometrics /based on measurements and data derived from direct measurement of a part of the human body0 include1

$inger!scan $acial Recognition "ris!scan Retina!scan 2and!scan

.ehavioral biometrics /based on

CAPTURING F I!AG" #$ %TANDARD &ID" CA!"RA%

%he image is optical in characteristics and may be thought of as a collection of a large number of bright and dark areas representing the picture details. At an instant there will be large number of picture details existing

measurements and data derived from an action0 include1

simultaneously each representing the level of brightness of the scene to be reproduced. "n other words the picture information would is a function of two of variables1 time and space. %herefore it re&uire infinite to number channels to transmit optical information corresponding simultaneously. picture are elements practical %here

An automated mechanism
that scans and captures a digital or an analog image of a module0 living personal characteristics./enrollment

Another entity which handles

compression, processing, storage and compression of the captured data with stored data /database0

difficulty in transmitting all information simultaneously so we use a method called scanning. 2ere the conversion of optical information to electrical form and its transmission is carried out element by element one at a time in a se&uential manner to cover the entire image. A %3 camera converts optical information into electrical information, the amplitude of which varies in accordance with variation of brightness.

%he third interfaces with the

application system / identification module0

Im'(ementation of Face Recognition Techno(ogy

%he implementation of face recognition technology include the following four stages1 data ac&uisition input processing face image classification and

decision making

Data ac)*isition
%he input can be recorded video of the speaker or a still image. A sample of 7 sec duration consists of a 8; frame video se&uence. <ore than one camera can be used to produce a 9=

Com'onents of Face Recognition %ystems

representation of the face and to protect against the usage of photographs to gain unauthori'ed access.

high!resolution search only after a head!

In'*t 'rocessing
A pre!processing module locates the eye position and takes care of the surrounding lighting condition and colour variance. $irst the presence of faces or face in a scene must be detected. >nce the face is detected, it must be locali'ed and normali'ation process may be re&uired to bring the dimensions of the live facial sample in alignment with the one on the template. $acial recognition software falls into a larger group of technologies known as biometrics. $acial recognition methods may vary, but they generally involve a series of steps that serve to capture, analy'e and compare your face to a database of stored images. 2ere is the basic process that is used by the $aceit system to capture and compare images1

like shape is detected.

A(ignment
>nce a face is detected, the system determines the head s position, si'e and pose. A face needs to be turned at least 9; degrees toward the camera for the system to register it.

Norma(i+ation
%he image of the head is scaled and rotated so that it can be registered and mapped into an appropriate si'e and pose. Normali'ation is performed regardless of the head s location and distance from the camera. ?ight does not impact the normali'ation process.

Re'resentation
%he system translates the facial data into a uni&ue code. %his coding process allows for easier

Detection
,hen the system is attached to a video surveillance system, the recognition software searches the field of view of a video camera for faces. "f there is a face in the view, it is detected within a fraction of a second. A multi!scale algorithm is used to search for faces in low resolution. /An algorithm is a program that provides a set of instructions to accomplish a specific task0. %he system switches to a

comparison of the newly ac&uired facial data to stored facial data.

!atching
%he newly ac&uired facial data is compared to the stored data and /ideally0 linked to at least one stored facial representation. %he heart of the $ace"t facial recognition system is the ?ocal $eature Analysis /?$A0 algorithm. %his is the mathematical techni&ue the

system uses to encode faces. %he system maps the face and creates a faceprint, a uni&ue numerical code for that face. >nce the system has stored a faceprint, it can compare it to the thousands or millions of @:!byte file. faceprints stored in a +sing facial recognition

False rejecti ! rates "FRR#

%he probability that a system will fail to identify an enrolee. "t is also called type 7 error rate FRR= NFR/NEIA ,here $RRB false reAection rates N$RB number of false reAection rates N)"AB number of enrolee identification attempt

database. )ach faceprint is stored as an software, police can 'oom in with cameras and take a snapshot of a face.

.ow of Face Recognition Wor/

3isionics, company based in a New Cersey is one of the many developers of facial recognition technology. %he twist to its particular software, $ace it is that it can pick

Performance
Fa(se acce'tance rate ,FAR%he probability that a system will incorrectly identify an individual or will fail to reAect an imposter. "t is also called as type 8 error rate. FAR= NFA/NIIA ,here $ARB false acceptance rate N$AB number of false acceptance N""AB number of imposter identification attempts

someone s face from the rest of the scene and compare it to a database full of stored images. "n order for this software to work, it has to know what a basic face looks like. $acial recognition software is based on the ability to first recogni'e faces, which is a technological feat in itself and then measure the various features of each face. "f you look at the mirror, you can see that your face has certain distinguishable landmarks. %hese are the peaks and valleys that make up the different facial features. 3isionics defines these landmarks as nodal points. %here are about @D nodal points on a

human face. 2ere are few nodal points that are measured by the software. distance between the eyes width of the nose depth of the eye socket cheekbones Aaw line chin %hese nodal points are measured to create a numerical code, a string of numbers that represents a face in the database. %his code is called faceprint. >nly 7: to 88 nodal points are needed for faceit software to complete the recognition process.

EA new proAect to address this is under way in (*9H G Allow for the specification of additional visible information discernable by an operator pertaining to the face, such as gender and eye color E%o improve identification performance G3erify that specified allow format for and good compression

facerecognition performance E%ested standard using leading algorithms on passport databases G.est practice appendices developed to allow for optimi'ed face recognition ENote1 some "(> best practices are "*A> re&uirements

Un0erstan0ing The Face Image Format %tan0ar0s

%he $ace "mage (tandard >n 9= $ace Recognition *onformance %esting (tandards and

Face Image Ty'es

$asic %he fundamental $ace "mage %ype that specifies a record format including header and image data. ENo mandatory scene, type. Fr !tal A face that adheres to additional re&uirements appropriate for frontal face recognition and#or human examination. )ither $ull or %oken. photographic and digital re&uirements are specified for this image

Genera( A''roach to the Face Format

(pecify face images because there is no agreement on a standard face recognition template E+nlike finger minutia F G (pecify how a photograph should appear rather than how to take the photograph /i.e. lighting and cameras0

%tan0ar0s Activities
"N*"%( <7 has a new work item to amend the AN(" 8= face standard to accommodate depth data.

A similar technical contribution has been made to (*9H#,I9, to be presented in (outh Africa this summer. )very biometric should have standardi'ed data formats, and in this context, 9= face is no different than hand geometry, voice, etc. %his does not imply that 9= face automatically will be put on passports or in next generation N"(% records as a piggyback to 8= face.

probability performance

of

good

biometric

$ocus, )xposure, )xpression, etc.

A0vantages
7.%here are many benefits to face recognition systems such as its convinence and social acceptability. All you need is your picture taken for it to work. 8.$ace recognition is easy to use

Performance Testing
E%ested 9= face subAects recognition in the on $ace cooperative

and in many cases it can be performed without a person even knowing. 9.$ace recognition is also one of the most inexpensive

Recognition Irand *hallenge /$RI*0

Conformance Testing

%tan0ar0s

an0

biometric in the market and its prices should continue to go down.

%tan0ar0s Activities
"N*"%( <7 is a new work item to study specify methods that automatically insure that face data complies with the face standards. %his new work item is currently being developed by "dentix and N"(%. *onformance is for the format /make sure bytes are in the right places0, photographic, digiti'ation, and scene

Disa0vantage
7. $ace recognition systems cant tell the difference between identical twins. APPLICATIONS %he natural use of face recognition technology is the replacement of P"N, physical tokens or both needed in automatic authori'ation or identification schemes. Additional uses are automation of human identification or role authentication in such cases where assistance of another human needed in verifying the "= cards

re&uirements. (cene re&uirements in the standard are to be evaluated using an automated i%a&e '(alit) % *(le.

Image 1*a(ity !o0*(e

E<easures scene information to determine conformance to standard and the

and its beholder. CONCLUSION $ace technologies have been recognition associated

generally with very costly top secure applications. %oday the core technologies have evolved and the cost of e&uipments is going down dramatically due to the intergration and the increasing processing power. *ertain application of face recognition technology are now cost effective, reliable and highly accurate. As a result there are no technological or financial barriers for stepping from the pilot proAect to widespread deployment.