Motion Compensated Deinterlacing With Efficient Artifact Detection for Digital Television Displays

Abstract

Deinterlacing is a scanning-format conversion technique that can convert interlaced pictures to progressive ones. Progressive televisions or monitors intended for displaying conventional interlaced programs have to apply deinterlacing techniques as required. Motion adaptive deinterlacing has been used widely in various kinds of commercial products, but the picture quality in motion areas still has much room for improvement. Motion compensated deinterlacing solves this problem by interpolation along the motion trajectory and thus is becoming the main trend of the next generation of deinterlacing algorithms. However, it suffers visible artifacts caused by incorrect motion vectors. To improve such imperfect pictures, this paper presents a motion compensation algorithm with more accurate motion estimation and more efficient artifact detection for digital television displays. Soft-switching between intra and inter interpolations is realized by characterizing different areas of the picture. Simulation results indicate that the proposed scheme can produce high-quality pictures with less flicker and imperceptible artifacts than a number of other techniques that were tested. In particular, the simulated peak signal-to-noise ratio (peak SNR) is basically 3-10 dB better than a comparable and prominent deinterlacing method.