Hasil Pencarian  ::  Simpan CSV :: Kembali

"In this paper, a 3-D face recognition system is developed usinga cylindrical structure of hidden layer neural network and its optimization through genetic algorithms. The cylindrical structure of hidden"

"Proses optimasi pada Probabilistic Neural Network (PNN) dapat dilakukan terhadap nilai smoothing parameter maupun struktur neuron. Setiap permasalahan memiliki nilai smoothing parameter optimal yang berbeda. Optimasi struktur neuron bertujuan untuk mereduksi banyak neuron yang digunakan sehingga dapat mempersingkat waktu komputasi. Skripsi ini membahas proses pencarian nilai smoothing parameter optimal menggunakan algoritma genetika dan struktur neuron optimal menggunakan algoritma ortogonal dalam sistem pengenal wajah. Terdapat dua jenis teknik optimasi yang akan dibahas, lalu membandingkan hasilnya dengan PNN struktur utuh dan backpropagation. Data wajah yang digunakan berupa foto infra merah dan cahaya tampak.

---

Optimization of Probabilistic Neural Network (PNN) can be performed to the value of smoothing parameter and neuron structure. Every problem has different value of smoothing parameter. Optimization of neuron structure aims to reduce the number of neurons used, in order to shorten computation time.

This thesis discusses the process of finding the optimal value of smoothing parameter using genetic algorithms and optimal neuron structure using orthogonal algorithms in face recognition system. Two types of optimization techniques which will be discussed, then the results are compared with full structure PNN and backpropagation. Face data used in the form of infrared and visible light images."

"ABSTRAKPengenalan wajah merupakan salah satu topik pada ilmu pengolahan citra yang sering dikembangkan. Salah satu dari metode pengenalan wajah ialah dengan menggunakan jaringan saraf tiruan. Jaringan saraf tiruan mengenali wajah-wajah dengan cara mempelajari wajah-wajah yang disediakan untuk pembelajaran. Metode pembelajaran yang digunakan pada tulisan ini ialah dengan lapisan tersembunyi berbentuk hemisfer yang merupakan pengembangan dari metodebackpropagation dengan data masukan yang direduksi oleh algoritma PCA. Metode ini menggunakan informasi sudut wajah pada citra sebagai parameter masukan selain data citra wajah tersebut. Seiring dengan majunya teknologi pengambilan gambar, metode ini dapat digunakan untuk mengenali wajah secara tiga dimensi.

---

ABSTRACTFace recognition is one of most discussed topics in image processing. A method used for face recognition is using artificial neural networks. Artificial neural network recognizes faces by learning the faces given to train. The learning method proposed in this paper is using hemispheric structure hidden layer which is an improvement of backpropagation algorithm with reduced data as input using principal component analysis algorithm. This method needs face’s angle on the image as parameter inputs instead of only face data. As the technology of capturing image growing, this method can be applied as an algorithm for 3D face recognition."

"[Dewasa ini, teknologi berkembang dengan sangat pesat, salah satu contoh teknologi yang sedang marak beberapa tahun belakangan ini adalah 3D face recognition. Teknologi ini menggabungkan data biometrik berupa wajah orang yang diambil dari beberapa sudut (horizontal dan vertikal) dan jaringan saraf tiruan. Untuk memperbaiki tingkat rekognisi yang rendah pada saat menggunakan data crisp, maka digunakanlah metode fuzzy. Percobaan akan dilakukan sebanyak tiga kali karena terdapat tiga cluster yang masing-masing cluster terdiri dari beberapa set orang. Pertama-tama, data akan diolah secara bertahap pada fase fuzzification dimulai dari parameter ekspresi, orang, dan sudut. Tahapan selanjutnya adalah membuat referensi pada fase fuzzy manifold untuk kemudian digunakan pada fase fuzzy nearest distance. Pada fase fuzzy nearest distance akan dicari jarak terpendek dari data testing dengan referensi yang sudah ada. Hasil keluaran dari sistem ini adalah kombinasi sudut horizontal dan vertikal dari tiap-tiap cluster yang nantinya akan dimasukkan kedalam Jaringan Saraf Tiruan (JST) dengan lapis tersembunyi berstruktur hemisfer untuk mendapatkan tingkat rekognisi. Secara keseluruhan rata-rata tingkat rekognisi setiap cluster sudah bisa mencapai 80%. Hal ini menunjukkan sistem sudah cukup optimal dalam mengenali pola wajah yang ada. ;The development of technology is growing rapidly, one of the examples of the technology that is emerging in recent years is 3D face recognition. This technology combines biometric data in form of faces which are taken from several angles (combination of horizontal and vertical angles) and artificial neural network. In order to improve the low recognition rate from crisp data, fuzzy method is used. The experiment will be performed three times because there are three cluster which are consist of several set of person. Firstly, the data will be processed step by step in fuzzification phase starting from the level of expression continued with the level of face and lastly is the level of person. With the use fuzzification, the crisp data can be converted into fuzzy. The next step is to make references in fuzzy manifold phase in order to be used in fuzzy nearest distance phase. In fuzzy nearest distance phase, the shortest distance between the testing data the references will be processed in artificial neural network with hemispheric structured hidden layer. Generally, the average of the all recognition rate can reach up to 80% which means that the system can recognize the face pattern quite good. ;The development of technology is growing rapidly, one of the examples of the technology that is emerging in recent years is 3D face recognition. This technology combines biometric data in form of faces which are taken from several angles (combination of horizontal and vertical angles) and artificial neural network. In order to improve the low recognition rate from crisp data, fuzzy method is used. The experiment will be performed three times because there are three cluster which are consist of several set of person. Firstly, the data will be processed step by step in fuzzification phase starting from the level of expression continued with the level of face and lastly is the level of person. With the use fuzzification, the crisp data can be converted into fuzzy. The next step is to make references in fuzzy manifold phase in order to be used in fuzzy nearest distance phase. In fuzzy nearest distance phase, the shortest distance between the testing data the references will be processed in artificial neural network with hemispheric structured hidden layer. Generally, the average of the all recognition rate can reach up to 80% which means that the system can recognize the face pattern quite good. , The development of technology is growing rapidly, one of the examples of the technology that is emerging in recent years is 3D face recognition. This technology combines biometric data in form of faces which are taken from several angles (combination of horizontal and vertical angles) and artificial neural network. In order to improve the low recognition rate from crisp data, fuzzy method is used. The experiment will be performed three times because there are three cluster which are consist of several set of person. Firstly, the data will be processed step by step in fuzzification phase starting from the level of expression continued with the level of face and lastly is the level of person. With the use fuzzification, the crisp data can be converted into fuzzy. The next step is to make references in fuzzy manifold phase in order to be used in fuzzy nearest distance phase. In fuzzy nearest distance phase, the shortest distance between the testing data the references will be processed in artificial neural network with hemispheric structured hidden layer. Generally, the average of the all recognition rate can reach up to 80% which means that the system can recognize the face pattern quite good. ]"

"Pengenalan wajah telah menjadi topik pengolahan citra yang banyak mengalami perkembangan. Pengembangan yang dilakukan bertujuan mengatasi kesulitan-kesulitan dalam pengenalan wajah, diantaranya pose pengambilan gambar. Penelitian ini bertujuan membuat rancang bangun rekognisi wajah tiga dimensi dengan sistem Fuzzy dan Jaringan Saraf Tiruan Hemisfer untuk mengatasi masalah tsersebut. Sistem fuzzy bertujuan untuk mengestimasi sudut pengambilan gambar dengan menempatkan informasi gambar ke suatu titik di ruang vektor fuzzy atau manifold dari data referensi menggunakan jarak terdekat fuzzy. Informasi sudut akan diteruskan ke jaringan saraf tiruan yang mengenali wajah-wajah dengan cara mempelajari wajah-wajah yang disediakan untuk pembelajaran. Informasi gambar yang dimasukkan ke dalam jaringan saraf tiruan terlebih dahulu dikompresi dengan metode Principle Component Analysis (PCA). Keunggulan jaringan saraf tiruan hemisfer dalam pengenalan wajah tiga dimensi adalah adanya faktor pengali neuron yang besarnya bergantung dari informasi sudut pengambilan gambar, sehingga gambar dua dimensi dapat diproyeksikan ke ruang tiga dimensi. Metode pembelajaran yang digunakan pada tulisan ini ialah pengembangan dari metode backpropagation. Penelitian diawali dengan pengambilan data dari alat pengambil gambar wajah tiga dimensi, perancangan sistem fuzzy dan jaringan saraf tiruan dalam MATLAB, dan validasi masing-masing sistem dengan data yang diambil. Sistem ini kemudian dikombinasikan dalam perangkat lunak MATLAB dan diuji dengan sebuah prototipe yang terdiri atas satu kamera. Hasil penelitian menunjukkan tingkat rekognisi sistem sebesar 76,29% pada saat validasi dan 37% saat aplikasi sistem satu kamera. Dari penelitian ini dibuktikan sistem dapat diaplikasikan untuk merekognisi wajah tiga dimensi, namun harus diperhatikan keakuratan pemotongan gambar untuk mendapat hasil yang akurat.

---

Face recognition is currently a highly discussed topic on image processing. The developments are aimed to overcome problems on recognizing face, such as various pose of image object. The study tries to solve the problem by creating a system design of 3D face recognition using a fuzzy system and Hemispheric Structure Hidden Layer of Artificial Neural Network to overcome the problem. The fuzzy system estimates pose information of the object taken. It is done by mapping the image taken to a point in a fuzzy vector space or manifold using fuzzy nearest distance. Pose information is then projected to the artificial neural network which is able to recognize faces after formerly learned a set of learning database. The data submitted to the artificial neural network is compressed by Principle Component Analysis (PCA). Main advantage of hemispheric neural network on 3D face recognition is the multiplying factor which values depend on the image pose information, so that the two dimensional images can be projected into three dimensional space. Learning method used in this study is an expansion of backpropagation. The study begins by taking experimental data from 3D face capturing devices, developing fuzzy system and artificial neural network in MATLAB, and validating both systems. The system is then combined in MATLAB and tested by a single unit camera prototype. Results show the system able to reach recognition rate of 76.29% on validation and 37% on single unit camera application. The study proves that the system is applicable for a 3D face recognition system, however the accuracy of image cropping should be taken into consideration for an accurate result."

"Dunia digital khususnya image processing berkembang seiring waktu berjalan dikarenakan kebutuhan masyarakat dan pentingnya keamanan sistem berbasis digital. Salah satu teknologi yang sangat mengalami kemajuan pesat adalah pengenalan wajah (face recognition) menggunakan artificial intelligence. Wajah seseorang yang sudah terdaftar di dalam database akan dikenali oleh sistem untuk keperluan validasi atau verifikasi. Di dalam penelitian ini dirancang sistem pengenalan wajah (face recognition) menggunakan algoritma machine learning dan Principal Component Analysis (PCA) sebagai pereduksi dimensi. Pengujian dilakukan dengan menggunakan beberapa metode, yakni: Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbour (K- NN), Logistic Regression (LR), Multi-Layer Perceptron (MLP) dan Convolutional Neural network (CNN). CNN berfokus pada layer dan tidak memerlukan reduksi dimensi, sehingga hasilnya lebih akurat. Model machine learning yang digunakan untuk classifier selain CNN adalah standar/default, sedangkan CNN menggunakan arsitektur LeNet-5, dengan dropout rate sebesar 0.25. Training dilakukan selama 60 epoch dengan loss function crosscategorical entropy, optimizer Adam, dan batch size sebesar 20. Data masukan adalah citra wajah berukuran 64 × 64 × 1 yang diperoleh dari dataset olivetti faces. Akurasi tertinggi metode PCA, SVM, maupun LR sebesar 91.25%, sementara akurasi terbaik CNN mencapai 98.75%. Selain akurasi, pemakaian confusion matrix dan classification report digunakan untuk menguji performa metode yang ada melalui evaluasi model klasifikasi.

---

The digital world, especially image processing, is evolving due to the needs of society and the importance of digital-based system security. One of the technologies that are rapidly progressing is face recognition using artificial intelligence. The system will recognize a person's face already registered in the database for validation or verification purposes. A face recognition system was designed using machine learning algorithms and Principal Component Analysis (PCA) as dimension reduction in this study. Testing is conducted using several methods: Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbour (K-NN), Logistic Regression (LR), Multi-Layer Perceptron (MLP) and Convolutional Neural network (CNN). CNN focuses on layers and does not require dimensional reduction to increase the accuracy of the result. The machine learning model used for classifiers other than CNN is standard/default settings, while CNN uses the LeNet-5 architecture, with a dropout rate of 0.25. The training was conducted for 60 epochs with loss function cross-categorical entropy, optimizer Adam, and batch size of 20. Input data is a 64 × 64 × 1 facial image obtained from the Olivetti faces database. The highest accuracy of PCA, SVM and LR methods was 91.25%, while CNN's best accuracy reached 98.75%. In addition to accuracy, the use of confusion matrix and classification report is used to test the performance of existing methods through the evaluation of classification models."

"Biometrik adalah salah satu teknologi cangih yang banyak dipakai untuk menjadi bagian dari sistem keamanan di berbagai bidang. Teknologi biometrik yang ada di sekitar kita ada berbagai macam seperti sistem identifikasi retina, iris mata, telapak tangan, sidik jari dan wajah. Banyak komputer atau laptop yang dilengkapi oleh kamera digital atau webcam yang terintegrasi dengan sistem yang ada di komputer itu sendiri. Teknologi camera digital semakin hari juga semakin canggih dalam pengambilan gambar yang dapat disesuaikan dengan situasi apapun contohnya seperti didalam ruangan atau diluar ruangan. Hal ini memungkinkan untuk pengaplikasian pngenalan wajah sebagai sistem autentikasi pengganti password selain fingerprint. Penelitian ini difokuskan pada perancangan aplikasi system pengenalan wajah menggabungkan dua metode yakni jaringan saraf tiruan dan metode pencocokan pola. Input dari sistem pengenalan wajah ini diambil dari webcam yang sudah melalui proses pre-processing dan sudah difokuskan ke bagian wajah dengan sistem pendeteksi wajah dengan metode pattern matching. Selain itu hasil dari pre-processing juga digunakan sebagai data training atau pelatihan. Gambar wajah hasil dari preprocessing ini kemudian masuk ke proses pengenalan menggunakan algoritma jaringan saraf tiruan. Hasil dari proses pengenalan wajah adalah berupa nama dari wajah orang yang dikenali. Sistem ini telah diuji pada lebih dari 36 sampel wajah yang diambil dari 12 orang.Hasil akhir menunjukan bahwa sistem ini berhasil mengidentifikasi sampel-sampel wajah tersebut dengan tingkat keberhasilan mencapai 86%.

---

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 preprocessing 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% success rate."

"Dalam beberapa tahun ini, telah banyak penelitian yang berhubungan dengan pengenalan pola dilakukan untuk mengidentifikasi wajah manusia secara otomatis. Skripsi ini membahas sistem pengenal wajah berbasis jaringan syaraf tiruan tunggal dan ensemble. Kamera infra merah digunakan pada penelitian untuk mengatasi masalah pencahayaan yang ditemui pada kamera visible light. Tahapan sistem terdiri dari pra-pemrosesan, ekstraksi fitur oleh PCA, pelatihan, dan pengujian. Pada percobaan dengan jaringan syaraf tunggal, algoritma backpropagation diterapkan dan sejumlah parameter divariasikan untuk mencapai performa yang optimal. Pada percobaan dengan jaringan syaraf ensemble, sejumlah skema digunakan antara lain: algoritma backpropagation, algoritma NCL, dan parsialisasi data dengan kedua algoritma tersebut. Hasil yang ditinjau dari recognition rate menunjukkan jaringan syaraf ensemble, yang terdiri dari sejumlah jaringan syaraf, memberikan performa yang lebih baik dibandingkan sebuah jaringan syaraf tunggal. Bila dioptimalkan, jaringan syaraf ensemble dapat menghasilkan recognition rate sebesar 99.9%.

---

People in pattern recognition have been working on automatic recognition of human faces for years. The focus of this thesis is a face recognition system based on both individual and ensemble neural network. An infrared camera is utilized to overcome the illumination matter encountered by visible light cameras. The procedure of the system consists of pre-processing, feature extraction by PCA, training and testing.

In the experiment of individual neural network, back-propagation algorithm is applied and some parameters are varied to obtain the optimum performance. In the experiment of ensemble neural network, some schemes such as back-propagation algorithm, NCL algorithm and partition of data using both algorithms are particularly examined.

According to the recognition rate, the results show that ensemble neural network, which is made up of several neural networks, have better performance than a single neural network does. An optimized ensemble neural network may reach up to 99.9% of recognition rate."