Abstract:
We developed pre-processing equipment for an 8K video codec with low-pass filtering and noise reduction functions to improve the efficiency of video coding for an advance...Show MoreNotes: This article was mistakenly omitted from the original submission to IEEE Xplore. It is now included as part of the conference record.
Metadata
Abstract:
We developed pre-processing equipment for an 8K video codec with low-pass filtering and noise reduction functions to improve the efficiency of video coding for an advanced digital terrestrial 8K UHDTV broadcasting system. These functions are implemented by novel proposed method based on the wavelet shrinkage technique. The proposed method features that two types of shrinkage functions {S1, S2} are applied to each coefficient in the frequency band obtained by waveletpacket decomposition. S1 performs only noise reduction. S2 performs both low-pass filtering and noise reduction. In S2, low-pass filtering is weak if coefficient level is high. On the other hand, low-pass filtering is strong if coefficient level is low. The blurring of visually noticeable strongedge components can be suppressed by using S2. Preprocessing equipment that implements the proposed method improves coding efficiency for real-time 8K encoded video because unnoticeable weak-edge components and unnecessary noise components are reduced through low-pass filtering and noise reduction. We conducted experiments to confirm the operation of the pre-processing equipment’s low-pass filtering and noise reduction functions. In addition, we examined how the pre-processing equipment and 8K video codec improved the coding image quality of the video including the reduction of compression artifact such as block distortion.
Notes: This article was mistakenly omitted from the original submission to IEEE Xplore. It is now included as part of the conference record.
Published in: 2018 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)
Date of Conference: 06-08 December 2018
Date Added to IEEE Xplore: 06 May 2019
ISBN Information:
Print on Demand(PoD) ISSN: 2162-7843