Deinterlacing using second order polynomial estimation

摘要

Deinterlacing has been an essential technology for modern video broadcasting services. A number of deinterlacing algorithms have been proposed such as intra or inter-field deinterlacing algorithms. Furthermore, deinterlacing algorithms that use neural networks also have shown promising performance. As broadcasting services with high resolution such as HD and UHD are becoming widely available, processing time and hardware complexity issues have become important. In this paper, we propose second order polynomial methods which can be substituted for neural network algorithms. We also applied the polynomial algorithm to deinterlacing problems. Experimental results showed that deinterlacing using second order polynomials was more efficient than using neural network algorithms in terms of PSNR and processing time.