KELOMPOK FARHAN RONY

: THE DEVELOPMENT OF SYSTEM IDENTIFICATION

BIOMETRIC BY FACE RECOGNITION USING
NEURAL NETWORK METHOD AND PATTERN
MATCHING METHOD

Biometric is one of the modern technology features that is used

mostly as a part of security system in many types of application. There are so
many biometric technology options this day such as retina identification, iris,
eye, hand, finger print, and face. Many computers such as laptop are
completed with digital camera or webcam which integrated with in the
system computer it self. Camera technology is getting more sophisticated in
nowadays in capturing image from many situations such as indoor or
outdoor environmental. This technology allows the possibility to develop
face recognition as an option to authentication system in computer, other
than the most popular fingerprint. This final project focuses on the design of
face identification application using combination of two methods, neural
network method and pattern matching method. The input of the system is
taken from face detection algorithm with pattern matching method on
webcam images which focused on human face area and already pass
preprocessing first. The digital images from pre- processing are also used as a
training data. The preprocessed image is then passed into the recognition
process using neural network algorithm. The result of the recognition process
is the person’s credential which in this case the name. This system has been
tested over 36 samples taken from 12 people. Result show that the system has
identified the samples with 86% successrate.

Key Word : Biometric, face, Identification, Detection, Neural Network Template Matching,
Pattern Matching, Webcam, digital image

Biometrics is one of the most widely used technologies for

become part of the security system in various fields. This biometrics is even
It has been used since ancient Egypt. One of biometric technology is

face recognition system. In addition biometric technology that is around us

there are various kinds such as the retinal identification system, iris, palms

and fingerprints.

Facial recognition lately more and more utilized, between

another for biometric recognition systems such as search and indexing systems

digital image database and digital video database, access control security system

restricted areas, video conferencing, and human interaction with computers.

Human interaction with computers becomes the background for the creation of ideas

separately build a security system with the authentication process

biometrics. The equipment is quite risky to invite crime

crimes such as theft, both physical theft and data theft.

With the facial recognition authentication system installed on

computer or labtop is expected to minimize the existence of no crime

the.

Besides that, sophisticated equipment such as cellphones, computers especially

Labtop is now widely equipped with a digital camera or webcam

which is integrated with the existing system on the computer itself. Technology

Digital cameras are increasingly sophisticated in taking pictures

which can be adapted to any situation for example like in the room

or outside the room. This also allows for the introduction application

face as a password replacement authentication system other than fingerprint. In the

this authentication process there is a face identification process to show

user identity.

The artificial neural network (ANN) method is an information progress system

with the ability to learn, remember and solve problems

based on the learning process provided. This system has a structure that is

resembles a human neural network. With this structure, ANN can

solve problems related to classification and pattern matching,

function approach, optimization, vector quantization, and data grouping. Duty -

the task can be completed efficiently by a computer where the computer can

complete arithmetic operations and logarithmic calculations faster [1].

This facial biometric system is easier to use from the user's side, because it isn't

requires physical contact that endangers the user. Apart from that facial biometrics

also more economical because it only uses a digital camera, which is on research

It uses a webcam, compared to other biometrics with much more scanning tools

expensive. With the neural network method or a supported neural network

with a simple image processing this research is expected to get results

which is more accurate, cheaper and faster on the introduction authentication system

face.

To realize the ideas arising from the background above, there is

several methods that can be used include methods such as neural networks
artificial (neural network).

The input of this authentication system is a digital image or image

taken from the webcam recording after face detection and then recognized

with the method of Artificial Neural Networks (ANN). The whole system consists of 4

stage, the first stage of face detection with pattern matching, the second stage ie

existing data trained (training) in accordance with the parameters

ANN which has been determined through testing, the third stage of the identification process or

introduction, and the final stage of evaluation of success rates.

Various methods have been tested and applied to detect

human face. Like the template matching method (Template Matching), Model

Hidden Markov Model (HMM), Face oxygen (Eigenface),

Artificial Neural Networks (ANN), Fuzzy logic,. The method chosen as the method

supporting this identification system is a combination of the Neural Network method

Knockoff (ANN) and pattern matching methods (Pattern Matching).

The combination of the two methods is a collaboration method on

during face detection and face recognition. Face detection at the time of capture

picture at the first time, both for training data and for data

image that will be recognized, using the template matching method or that

also called pattern matching. Meanwhile facial recognition is done

during the identification process using the neural network method with

back propagation algorithm. Artificial Neural Network (ANN), Fuzzy logic,. The method chosen as the method

supporting this identification system is a combination of the Neural Network method

Knockoff (ANN) and pattern matching methods (Pattern Matching).

The combination of the two methods is a collaboration method on

during face detection and face recognition. Face detection at the time of capture

picture at the first time, both for training data and for data

image that will be recognized, using the template matching method or that

also called pattern matching. Meanwhile facial recognition is done

during the identification process using the neural network method with

back propagation algorithm.

Biometrics

Biometric technology is security technology that uses parts body or habits that come from our body as identity. Medical world say that there are parts of our body that

are also very unique habits which he created, like a signature. Important, not approved by more than one individual. Examples of body parts such as fingerprints or retina

of the eye. Though eye shape or color can be the same, but the eye's retina is not necessarily the same. Likewise with the voice and facial structure. These unique parts

are what then developed as a safety attribute. As part of security technology, biometrics has two functions which can be run separately. The first as a registrar ID or as a

verification tool (password). Biometric technology can only be applied anywhere. Start to Like certain items from unwanted access, like computer. For a crowded room

from people For example, to discuss about experiments or other criminals at airport. The way this security technology works is almost the same other security

technology that is very difficult for sensors. Sensor which

digunakan pada teknologi biometrik cenderung mahal dan semakin akurat

ketajamannya maka akan semakin mahal.

Selain sensor, bagian yang tidak kalah penting dari biometrik adalah data.

Bagaimana Anda menyimpan data pada sebuah sistem sangat penting. Sebab

biometrik adalah teknologi yang bergantung kepada data. Bila data yang disimpan

tidak aman atau lengkap, kemungkinan adanya penyusup ke system ini akan lebih

besar.

Menurut sistemnya biometrik sendiri terbagi atas tiga macam, sebagai

berikut:

a. Sistem yang menyimpan data langsung pada alat.

Dengan sistem ini, data akan disimpan pada media penyimpanan yang

berada dalam alat detektor. Jika sewaktu-waktu mesin harus di-reset atau

dikembalikan ke posisi awal, maka data yang ada dapat saja ikut terhapus.

Sehingga petugas harus meng-input ulang. Jika data yang dimasukkan

sangat banyak tentu akan sangat merepotkan, lain halnya bila data tidak

terlalu banyak. Biometrik dengan sistem ini sangat cocok untuk diterapkan

pada sebuah alat tertentu yang tidak digunakan oleh banyak orang atau

untuk untuk melindungi sebuah ruang khusus, yang juga tidak diakses oleh
banyak pengunjungnya.

b. Sistem yang menyimpan data pada jaringan.

Sistem yang kedua memanfaatkan jaringan untuk menyimpan datanya.

Sistem yang kedua sangat efektif bagi aplikasi yang memang

dipergunakan untuk banyak user. Misalnya saja untuk data absen

karyawan atau siswa. Bentuk fisik yang ditampilkan oleh alat juga tidak

perlu terlalu besar. Karena data tidak akan diproses langsung pada alat.

Melainkan dikirim dahulu ke sebuah jaringan baru kemudian diproses dan

disimpan. Sistem ini memang membutuhkan waktu lama. Tetapi cukup

efektif untuk data yang besar. Karena tidak akan terkena risiko data hilang

pada saat proses reset pada alat harus dilakukan.

c. Sistem yang menyimpan data pada sebuah chip.

Sistem yang terakhir ini menggunakan media tambahan berupa chip untuk

menyimpan data si pemilik ID. Sehingga untuk menggunakannya