期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：3
DOI：10.15680/ijircce.2015.0303016
出版社：S&S Publications
摘要：Image mining is defined as the discovery of image patterns in a given collection of images. It is an effortthat fundamentally draws upon knowledge in computer vision, image processing, data mining, machine learning,database, and artificial intelligence. Facial recognition helps to analyze and compare the patterns from the facialimages. Facial feature extraction is an automatic recognition of human faces by detecting its features i.e. eyes,eyebrows and lips. In this research work, features are extracted from the human facial images by using the existingFace Part Detection (FPD) algorithm and the newly proposed Gray Level Co-occurrence Matrix (GLCM) algorithm.FPD uses bounding box method and GLCM uses affine moment invariants method. Performance factors applied hereare feature extraction accuracy and execution time. The implementation of this work is performed in MATLAB 7.0.Based on the experimental results, it is observed that the proposed GLCM algorithm extracted the features moreaccurately with minimum execution time than FPD algorithm.
关键词：Feature Extraction; GLCM; FPD; Convolution; Bounding Box; Affine Invariant Moments; Binary;converter and Gabor features
