Research on the Algorithm of Adaptive Loop Filter Based on Flexible Quadtree

Article Preview

Abstract:

The algorithm of adaptive loop filter based on flexible quad tree (Q-ALF) is put forward targeting the problem that deblocking loop filter used by H.264/AVC cant further improve video coding quality. This algorithm, through the original image and the image after deblocking filter, works out wiener filter and selects a wiener filter with optimal size by taking the SAD of the image after Wiener filtering and the original image as RDO cost and based on RDO model, then uses this filter to filtrate the area that needs filtration. Simulation result shows that Q-ALF algorithm can improve the PSNR of the reconstructed image and decrease bit rate; meanwhile, Q-ALF algorithm can better eliminate blocking effect, improving the subjective performance of the reconstructed image.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

2491-2495

Citation:

Online since:

September 2013

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2013 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] TAN Lili: The Hardware Design and FPGA Verification of Adaptive Loop Filter for H. 264, Southeast University, (2010).

Google Scholar

[2] LI Hongwei: Research on several key technology of HEVC, Xidian University, (2012).

Google Scholar

[3] CAI Xiaoxia, CUI Yansong, DENG Zhongliang, e tal: Model of Next-generatiOn Video Standard and Relative Key Technologies, ideo Engineering. Forum Vol. 36 (2012) 80-84.

Google Scholar

[4] LI Weiwei. Research on In-Loop Filter Algorithm for H. 264/AVC, Xidian University, (2007).

Google Scholar

[5] LIU Yu, HU Xiaoshuan, DUAN Jizhong, e tal. Algorithm Analysis and Comparison of the New Video Coding TechnOlogy HEVC. ideo Engineering, Forum Vol. 36 (2012)45-49.

Google Scholar

[6] WANG Gang, CHEN He-xin, CHEN Mian-shu, e tal: An improved filtering algorithm for H. 264 fractional-pel interpolation. Journal of Jilin University (Engineering and Technology Edition) , Forum Vol. 43(2013)173.

Google Scholar