Noise robust motion refinement for motion compensated noise reduction

被引：2

作者：

Kim, Jong-Sun ^{[1
]}

Kim, Lee-Sup ^{[1
]}

机构：

[1] Korea Adv Inst Sci & Technol, Seoul, South Korea

来源：

IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS | 2008年 / E91D卷 / 05期

关键词：

motion refinement; noise reduction; motion estimation; motion compensated noise reduction;

D O I：

10.1093/ietisy/e91-d.5.1581

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A motion refinement algorithm is proposed to enhance motion compensated noise reduction (MCNR) efficiency. Instead of the vector with minimum distortion, the vector with minimum distance from motion vectors of neighboring blocks is selected as the best motion vector among vectors which have distortion values within the range set by noise level. This motion refinement finds more accurate motion vectors in the noisy sequences. The MCNR with the proposed algorithm maintains the details of an,image sequence very well without blurring and joggling. And it achieves 10% bit-usage reduction or 0.5 dB objective quality enhancement in subsequent video coding.

引用

页码：1581 / 1583

页数：3

共 7 条

[1] NOISE-REDUCTION FILTERS FOR DYNAMIC IMAGE SEQUENCES - A REVIEW [J].

BRAILEAN, JC ;

KLEIHORST, RP ;

EFSTRATIADIS, S ;

KATSAGGELOS, AK ;

LAGENDIJK, RL .

PROCEEDINGS OF THE IEEE, 1995, 83 (09) :1272-1292

[2]

CHEONG HY, 2004, ICIP 04 2004 INT C O, V2, P965

[3]

*ISO IEC, 2003, JTC1SC29WG11NO400 IS

[4]

OLIVIERI S, 1999, 1999 IEEE 3 WORKSH S, V2, P345

[5] Estimation of image noise variance [J].

Rank, K ;

Lendl, M ;

Unbehauen, R .

IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING, 1999, 146 (02) :80-84

[6]

*SAM EL CO LTD, 2006, S5L3210 SAMS EL CO L

[7]

Zlokolica V, 2004, IEEE IMAGE PROC, P1465

← 1 →