Wavelet Transform Based Low Pass Filters and Interpolation Filters in Digital Image Communication Systems

디지털 영상 통신 시스템에서 웨이블릿 변환 기반 저역 필터와 보간 필터

  • 유훈 (동서대학교 컴퓨터정보공학부)
  • Published : 2006.04.01

Abstract

In digital image communication systems, YUV 4:2:0 or YUV 4:2:2 images filtered and subsampled versions of YUV 4:4:4 images are utilized and these images recover their size by an interpolation filter. Low pass filters and interpolation filters in the image communication systems are generally utilized. Thus, to improve the image quality, efficient low pass filters and interpolation filters are still required. In this paper, we propose new and efficient low pass filters and interpolation filters and their design method. The low pass filters and interpolation filters used in the MPEG-2 system were developed independently. We utilize wavelet transforms to jointly design low pass filters and interpolation filters. Simulation results show that the proposed filters are superior to the filters used in MPEG-2 in terms of PSNR. In addition, the length of the proposed interpolation filters is shorter than that of the filters used in the MPEG 2 system.