Adaptive Threshold Edge Detection Algorithm Based on APDCSF and Anti-symmetrical Biorthogonal Wavelet Transform

Abstract

This paper proposes an edge detection algorithm based on adaptive threshold which is obtained by histogram matching edge gradient value. The implementation of this algorithm is made up of three procedures: all phase discrete cosine sequency filtering, wavelet decomposition and adaptive threshold processing. The image is decomposed into low frequency and high frequency subbands by using all phase discrete cosine sequency filter (APDCSF). Experimental results on test images reveal that compared with the edge detection algorithm based on artificial threshold, the proposed one can create dynamic threshold adaptively and show better edge detection performance especially in terms of edge positioning accuracy.