No abstract

No abstract

No abstract

No abstract

ABSTRACT:This paper provides an insight into the effectiveness of edge detection algorithms as a method to enhance the performance of a correlation filter in face recognition application. Correlation filters are a class of pattern classifiers that makes the classification based on the Fourier magnitude spectrum of the images. CMU-PIE expression set is used or the experiments. We found that edge detection algorithms enhance the performance of correlation filters considerably. This was also established by the mean entropy scores of edge detected images. The reason for the improvement in the performance would be the narrowing of the magnitude spectrum spread in edge detected face images.