ABSTRACT
ABSTRACT: The use of phase congruency for marking features has significant advantages over gradient-based methods. It is a dimensionless quantity that is invariant to changes in image brightness or contrast, hence it provides an absolute measure of the significance of feature points. In this paper, identity recognition based on improved Log-Gabor phase congruency is proposed. The improved phase congruency algorithm is based on the improved local energy calculation method, frequency spread, and noise compensation tactics. The feature of improved phase congruency is of good location and recognition, and then the LDA method is used to project features into a low-dimensional space. The nearest neighbor classifier based on Euclidean distance is tested in the CASIA gait database. The experimental results show that our approach outperforms other state of the art automatic algorithms in terms of recognition accuracy.
