结合相位变换形状形成特性的多描述编码

doi:10.13190/jbupt.200804.107.025

北京邮电大学学报 ›› 2008, Vol. 31 ›› Issue (4): 107-111.doi: 10.13190/jbupt.200804.107.025

结合相位变换形状形成特性的多描述编码

马丽红1，2,于栋1,成桂锋1，2，3,卢汉清4

1.华南理工大学电子工程系，广州 510641；2.华南理工大学广东省计算机网络重点实验室，广州 510641；
3. 广东北电研发中心，广州 510220；4. 中国科学院国家模式识别重点实验室，北京 100080

收稿日期:2007-12-27 修回日期:1900-01-01 出版日期:2008-08-30 发布日期:2008-08-30
通讯作者: 马丽红

Multiple Description Coding Based on Shaping from Phase-Only Transform

MA Li-hong 1，2 , YU Dong 1。2 ,CHENG Gui-feng 1，2，3 ,LU Han-qing 4

（1. Department of Electronic, South China University of Technology, Guangzhou 510640, China;2. Guangdong Key Laboratory of Computer Network, South China University of Technology, Guangzhou 510640，China;
3. Guangdong Nortel Research and Development Center, Guangzhou, 510220,China;
4. National Laboratory of Pattern Recognition, Chinese Academy of Science, Beijing 100080, China

Received:2007-12-27 Revised:1900-01-01 Online:2008-08-30 Published:2008-08-30
Contact: MA Li-hong

摘要/Abstract

摘要：

提出一种利用相位变换形状保持特性的错误恢复编码新方法。编码结果的2个描述分别由相位图像和像素交织产生。该方法在有较大失真的情况下，可由相位图和失真幅值合成逼真的原图像：若2个描述都被正确接收，从相位图像和失真较少的幅度可重构一幅高质量的图像；若其中一个描述丢失，甚至在图像传输过程遭到严重破坏时，利用相同方法，能重构一幅可接受图像。实验结果显示，与像素交织和相扰的多描述算法相比，该算法的峰值信噪比（PSNR）可以高出1.5～3 dB。

关键词: 多描述编码, 相位变换, 像素交织, 相扰

Abstract:

We propose a new method for error correction coding based on shape preserving characteristic of the phase-only transform. It creates 2 descriptions by phase mapping and pixel interleaving。It synthesizes the image with only a little distortion by phases and damaged magnitudes. The receiver decodes a high quality image when both descriptions are received. Images, badly damaged due to the high error rate in transmission, can also be recovered from the only available description with an acceptable quality. Simulations show that it outperforms schemes such as pixel interleaving and phase scrambling by about 1.5~3 dB in peak signal to noise ratio (PSNR).

Key words: font-size: 10.5pt" lang="EN-US">multiple description codingmso-ascii-font-family: 'Times New Roman', mso-hansi-font-family: 'Times New Roman', mso-bidi-font-family: 'Times New Roman'">；font-size: 10.5pt" lang="EN-US">phase-only transformmso-ascii-font-family: 'Times New Roman', mso-hansi-font-family: 'Times New Roman', mso-bidi-font-family: 'Times New Roman'">；font-size: 10.5pt" lang="EN-US">pixel interleavingmso-ascii-font-family: 'Times New Roman', mso-hansi-font-family: 'Times New Roman', mso-bidi-font-family: 'Times New Roman'">；font-size: 10.5pt" lang="EN-US">phase scrambling

中图分类号:

TP391.11

马丽红,于栋,成桂锋,卢汉清. 结合相位变换形状形成特性的多描述编码[J]. 北京邮电大学学报, 2008, 31(4): 107-111.

MA Li-hong 1，2,YU Dong 1。2,CHENG Gui-feng 1，2，3,LU Han-qing 4. Multiple Description Coding Based on Shaping from Phase-Only Transform[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2008, 31(4): 107-111.

参考文献

[1] 张洋，张楠，尹宝才. 多描述编码研究现状[J]. 计算机学报，2007, 30(9):1612-1624.
Zhang Yang, Zhang Nan, Yin Baocai. Overview or resear- ches on mutiple description coding [J]. Chinese Journal of Computer, 2007, 30(9):1612-1624.
[2] Wang Y, Orchard M, Reibman A. Mutiple description coding using pairwise CT [J]. IEEE Trans Image Proc, 2001,10(3):351-366.
[3] Goyal V, Kovaˇcevic J. Generalized multiple description coding via CT[J].I EEE Trans Info Theory, 2001, 47(6): 2199-2224.
[4] 计文平，沈兰荪. 基于Mojette变换的多描述编码[J]. 电子学报，2007，35(3):526-529.
Ji Wenping, Shen Laisun. Multiple description coding by Mojettee transform[J].Acta Ele Sinica,2007,35(3):526-529.
[5] Chung Doe-Man, Wang Yao. Multiple description image coding using signal decomposition and reconstruction based on lapped orthogonal transforms [J]. IEEE Trans
Circuits and Sys for Video Tech, 1999, 9(6): 895–908.
[6] 张楠，吕岩，吴枫，等. 基于方向提升小波变换的多描述图像编码[J]. 自动化学报，2007, 33(6):567-576.
Zhang Nan, Lu Yan, Wu Feng, et al. Multiple deascription image coding baed on directional lifting wavelet[J]. Acta Auto Sinica, 2007,33(6):567-576.
[7] 陈婧，蔡灿辉，丁润涛. X树非平衡保护多描述编码[J].电子与信息学报，2007，35(3):1973-1977.
Chen Jin, Cai Canhui, Ding Runtao. X-tree unequal protected multiple description cod. [J]. Elec Info Tech 2007,35(3): 1973-1977.
[8] Shirani S, Gallant M, Kossentini F. Multiple description image coding using pre & post processing[C]//IEEE ICITC & C 2001. LasVegas, USA:IEEE Explore,2001:35–39.
[9] Olmo G, Tillo T. Directional multiple description scheme for still images [C]// ICECS 2003. Torino: IEEE Explore, 2003:886–889.
[10] Sadri K F, Shirani S. Multiple description coding of images using phase scrambling[C]// ICASSP04. HongKong：IEEE Explore, 2004: 41-44.
[11] Gharavi H. Multichannel mobile ad hoc links for multimedia comm[J]. Proc of IEEE, 2008,96(1):77–96.
[12] 杨戈,廖建新,朱晓民,王纯. 面向对等网络的流媒体接纳控制[J].北京邮电大学学报,2008,31 (1):48-51.
Yang Ge,Liao Jianxin,Zhu Xiaomin,Wang Chun. Peer-to- Peer Oriented Admission Control for Streaming Media[J].J. Beijing Univ. of Posts and Telecom ,2005,28 (4):87-90.
[13] 黄文辉,陈仁雷,张家谋. 数字视频图像质量客观测量方法改进与实现[J].北京邮电大学学报,2005,28 (4):87-90.
Huang Wenhui, Chen Renglei, Zhang Jimo. Improvementand implementation of objective digital video quality measurement[J].J.Beijing Univ. of Posts and Telecom, 2005,28 (4):87-90.
[14] Tillo T, Grangetto M, Olmo G. Redundant slice optimal allocation for H.264 multiple description coding[J]. IEEE Trans Circuits & Sys for Video Tech, 2008, 18(1): 59-70.

结合相位变换形状形成特性的多描述编码

Multiple Description Coding Based on Shaping from Phase-Only Transform

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价