期刊名称：International Journal of Computer Science & Information Technology (IJCSIT)
印刷版ISSN：0975-4660
电子版ISSN：0975-3826
出版年度：2015
卷号：7
期号：4
页码：93
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Human face expression is one of the cognitive activity or attribute to deliver the opinions to others. Thispaper mainly delivers the performance of appearance based holistic approach subspace methods based onPrincipal Component Analysis (PCA). In this work texture features are extracted from face images usingGabor filter. It was observed that extracted texture feature vector space has higher dimensional and hasmore number of redundant contents. Hence training, testing and classification time becomes more. Theexpression recognition accuracy rate is also reduced. To overcome this problem Symmetrical Weighted2DPCA (SW2DPCA) subspace method is introduced. Extracted feature vector space is projected in tosubspace by using SW2DPCA method. By implementing weighted principles on odd and even symmetricaldecomposition space of training samples sets proposed method have been formed. Conventional PCA and2DPCA method yields less recognition rate due to larger variations in expressions and light due to morenumber of feature space redundant variants. Proposed SW2DPCA method optimizes this problem byreducing redundant contents and discarding unequal variants. In this work a well known JAFFE databasesis used for experiments and tested with proposed SW2DPCA algorithm. From the experimental results itwas found that facial recognition accuracy rate of GF+SW2DPCA based feature fusion subspace methodhas been increased to 95.24% compared to 2DPCA method.
