外文翻译----一种基于嵌入式零树小波算法的鲁棒图像压缩新方法

《外文翻译----一种基于嵌入式零树小波算法的鲁棒图像压缩新方法》由会员分享，可在线阅读，更多相关《外文翻译----一种基于嵌入式零树小波算法的鲁棒图像压缩新方法（9页珍藏版）》请在毕设资料网上搜索。

1、A New Method of Robust Image Compression Based on the Embedded Zero tree Wavelet Algorithm Charles D. Creusere AbstractWe propose here a wavelet-based image compression algorithmthat achieves robustness to transmission errors by partitioningthe transform coefficients into groups and independently pr

2、ocessing eachgroup using an embedded coder. Thus, a bit error in one group does not affect the others, allowing more uncorrupted information to reach thedecoder. Index TermsCoefficient partitioning, embedded bitstream, error resilience,image compression, low complexity, wavelets. I. INTRODUCTION Rec

3、ently, the proliferation of wireless services and the internetalong with consumer demand for multimedia products has spurredinterest in the transmission of image and video data over noisycommunications channels whose capacities vary with time. In suchapplications, it can be advantageous to combine t

4、he source andchannel coding (i.e., compression and error correction) processesfrom both a complexity and an information theory standpoint .In this work, we introduce a form oflow-complexity joint sourcechannelcoding in which varying amounts of transmission errorrobustness can be built directly into

5、an embedded bit stream. Theapproach taken here modifies Shapiros embedded zerotree wavelet(EZW) image compression algorithm, but the basic idea canbe easily applied to other 1 wavelet-based embedded coders such This paper is organized as follows. In Section II, we discuss theconventional EZW image c

6、ompression algorithm and its resistanceto transmission errors. Next, Section III develops our new, robustcoder and explores the options associated with its implementation. InSection IV, we analyze the performance of the robust algorithm in thepresence of channel errors, and we use the results of thi

7、s analysis toperform comparisons in Section V. Finally, implementation and complexityissues are discussed in Section VI, followed by conclusionsin Section VII. II. EZW IMAGE COMPRESSION After performing a wavelet transform on the input image, the EZWencoder progressively quantizes the coefficients u

8、sing a form of bitplane coding to create an embedded representation of the imagei.e.,a representation in which a high resolution image also containsall coarser resolutions. This bit plane coding is accomplished bycomparing the magnitudes of the wavelet coefficients to a thresholdT to determine which

9、 of them are significant: if the magnitudeis greater than T, that coefficient is significant. As the scanningprogresses from low to high spatial frequencies, a 2-b symbol isused to encode the sign and position of all significant coefficients.This symbol can be a + or - indicating the sign of the sig

10、nificantcoefficient； a “0” indicating that the coefficient is insignificant； ora zerotree root (ZTR) indicating that the coefficient is insignificantalong with 2 all of the finer resolution coefficients corresponding tothe same spatial region. The inclusion of the ZTR symbol greatlyincreases the cod

11、ing efficiency because it allows the encoder toexploit interscale correlations that have been observed in most images. After computing the “significance map” symbols for a givenbit plane, resolution enhancement bits must be transmitted for allsignificant coefficients； in our implementation, we conca

12、tenate twoof these to form a symbol. Prior to transmission, the significance andresolution enhancement symbols are arithmetically encoded using thesimple adaptive model described in with a four symbol alphabet(plus one stop symbol). The threshold T is then divided by two, andthe scanning process is

13、repeated until some rate or distortion targetis met. At this point, the stop symbol is transmitted. The decoder,on the other hand, simply accepts the bitstream coming from theencoder, arithmetically decodes it, and progressively builds up thesignificance map and enhancement list in the exact same wa

14、y as theywere created by the encoder. The embedded nature of the bitstream produced by this encoderprovides a certain degree of error protection. Specifically, all of theinformation which arrives before the first bit error occurs can beused to reconstruct the image； everything that arrives after is lost.This is in direct contrast to many compression algorithms wherea single error can irreparably damage the image. Furthermore, wehave found that the EZW algorithm can actually detect an errorwhen its arithmetic decoder terminates (by decoding a stop