文章基本信息

标题：FACE LOCALIZATION-BASED TEMPLATE MATCHING APPROACH USING NEW SIMILARITY MEASUREMENTS
本地全文：下载
作者：GHASSAN MARWAN ABDULFATTAH ; MOHAMMAD NAZIR AHMAD
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：57
期号：3
出版社：Journal of Theoretical and Applied
摘要：In this paper, a number of similarity measurements have been developed, namely: Sum of Absolute Difference (OSAD), Sum of Square Difference (SSD) and Normalized Cross Correlation (NCC) in order to measure the correlation between the input image and the template image. In addition, two metrics were proposed, specifically: Sum Square T-distribution Normalized (SSTN) and Chi-Square distribution (Chi2) by which to measure matching between the two images. The result showed that optimized measurements overcome any drawbacks of NCC. Moreover, our results show SSTN and Chi2 as having the highest performance compared with other measurements. Sets of faces including: Yale, MIT-CBCL, BioID, Indian and Caltech were used to evaluate our techniques with success localization accuracy of up to 100%.
关键词：Face Localization; Template Matching; Similarity Measurements