Fine directional de-interlacing algorithm using modified sobel operation

Abstract

This paper presents an intra field de-interlacing algorithm for spatial edge preserving using the detection of accurate edge direction. Conventional intra field de-interlacing algorithms determine the edge direction at the pixel or half pixel level, so that they can be highly sensitive to noise and lead to image degradation. In this paper, the proposed algorithm first considers the edge tendency in the edge region, and then Candidate Direction Vectors (CDVs) are selected using the modified Sobel operation through the edge tendency. Finally, the CDVs are adaptively applied to interpolate the lost pixel in the edge region. Experimental results show that the proposed algorithm performs well with a variety of still and moving images compared with conventional intra field algorithms in the literature.