Edge-Based 3D Y/C Separation Using Wavelet Transform and Motion Compensation

Abstract

We propose new 3D adaptive Y/C separation algorithm for the composite video signal decoding. The exact calculation of the similarity is one of the key issues for the Y/C separation. We propose new approaches for the composite video signal decoding: 1) the use of the wavelet transform of the composite video signal for effective similarity computation, 2) an edge-based separation by using the composite video signal directly for sharp-edged results, 3) a simple motion-compensated similarity computation between frames. Also, the proposed algorithm has an adaptive architecture with which multiplication coefficients are adjusted by the similarity values between two pixels in lines of a field, two consecutive fields, and two consecutive frames. In case of comb failure, the proposed scheme resorts to the traditional band-pass filtering method. The experimental results show that the proposed method provides better performance than that of the conventional Y/C separation algorithms.