Artificial Intelligent Systems and Machine Learning

C. Anand ¹, R. Lawrance ¹

Affiliations
1 PG Department of Computer Applications, Ayya Nadar Janaki Ammal College, Sivakasi-626 124, Tamil Nadu, IN

Abstract

Face Recognition stands high as a significant research area since it has plenty of application domains in pattern recognition, image processing, biometrics etc. Researchers contributed lot of algorithms and techniques to uncover the mask of face recognition arena. In this paper, a Left-Right Hidden Markov Models (HMM) based face recognition algorithm along with Singular Value Decomposition (SVD) Coefficients is discussed. Human face is divided into seven facial regions and a small number of quantized SVD Coefficients were trained to choose the facial features. Order Statistic Filtering is used as a preprocessing operation for efficient computation. Using SVD Coefficients, a face is considered as a numerical sequence representing block of images which can be easily modeled by discrete HMM. The system is tested on Olivetti Research Laboratory (ORL) face database consist of 400 images of 40 persons in .pgm format. For training, five face images of a person are considered and our proposed system achieves a recognition rate of 96.5% with a computational speed of 0.22 seconds per image. The experimental results reveal that our proposed system outperforms many of the traditional face recognition methods tested on ORL database.

Keywords

Face Recognition, Hidden Markov Models (HMM), Order Statistic Filtering, Pattern Recognition, Singular Value Decomposition (SVD) Coefficients.

Full Text

     

M. Lydia Packiam Mettilda ¹, R. Lawrance ², R. Kalaivani ³

Affiliations
1 Department of Computer Application, Ayya Nadar Janaki Ammal College, Sivakasi, IN
2 Computer Application Department, Ayya Nadar Janaki Ammal College, Sivakasi, IN
3 Department of Computer Science and Information Technology, Ayya Nadar Janaki Ammal College, Sivakasi, IN

Abstract

With the advancement in technology, one thing that concerns the world and especially in the developing countries is the tremendous increase in population. With such a rapid rate of increase, it is becoming difficult to recognize each and every person because we have to keep up photos either in digital or hard copy format of every person at different time periods of his/her life. Sometimes database has the required information of that particular person, but it’s of no use as it is now obsolete. Deciding age of a person from digital photography is an intriguing problem. Age changes cause many variations in visible of human faces. Many aspects affect the appearance of a person’s face during the process of growing older. The aging process will explain with many factors such as health, living style, living place and weather condition etc. Face is a non-intrusive recognition, without user co-ordination able to recognize the person. Age classification system is generally composed of feature extraction and classification. It is used to estimate the age of a person from his/her face features. For the aging feature extraction, face images interpreted as decomposition of optimized wavelet transform with 49 feature vectors using Daubechies wavelet and the classifier of supervised neural network to discriminate the ranges of ages. The work is to classify the age range into child (1-10), teenage (11-20) young (21-30), middle aged (31-50) and old (51 and above).

Keywords

ASM, Wavelet, Age Classification, Neural Network.

Full Text

     

R. Kalaivani ¹, C. Devi Arockia Vanitha ², R. Lawrance ³, M. Lydia Packiam Mettilda ⁴

Affiliations
1 Department of Computer Science & Information Technology, Ayya Nadar Janaki Ammal College, Sivakasi 626 124, Tamilnadu, IN
2 Computer Science Department, The S.F.R. College for Women, Sivakasi 626123, Tamilnadu, IN
3 MCA Department, Ayya Nadar Janaki Ammal College, Sivakasi 626 124, Tamilnadu, IN
4 Department of Computer Application, ANJA College, Sivakasi 626 124, Tamilnadu, IN

Abstract

Most of the real world problems in engineering, medicine, industry, science and business also involve data classification. Classification is a supervised machine learning technique used to predict group membership for data instances. Pattern classification problems belong to the category of supervised learning. Pattern Classification involves assigning a label to a given input data. Neural Networks are an effective tool in the field of pattern classification, using training and testing data to build a model. Training neural networks in classification problems, especially when biological data are is a very challenging task. The protein superfamily classification problem, which consists of determining the superfamily membership of a given unknown protein sequence, is very important for a biologist for many practical reasons, such as drug discovery, prediction of molecular function and medical diagnosis. The objective of this work is creating a classification model for classifying data using Multilayer feed forward network. It contain two phases. First, classifier model was build for iris plant classification. Second, classifier model was build for protein sequence classification to know the organism of protein and family of the given protein sequence.

Keywords

Data Mining, Pattern Classification, Neural Network, Back Propagation, Iris Plant, Bioinformatics, Protein Sequence.

Full Text