Font Size: a A A

The Wavelet Transform-based Still Image Coding Technology Research

Posted on:2007-03-12Degree:MasterType:Thesis
Country:ChinaCandidate:H B DuFull Text:PDF
GTID:2208360185983407Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
This article makes researchs on the still image compression and coding with the task background of gathering military informations on sea. Because the boat on sea bumps with the heavy wind and surf, point to point microwave communication becomes difficult. The ultrashort communication is also can't be used for it's low travelling ability. But the shortwave communication becomes a valid image transmission choice for its strong travelling ability. On account of the shortwave's limited bandwidth, it is a key problem in image data processing that how to compress the image. The definition of image compression is the processing of decreasing original data in the position of insuring the image quality. The early research on still image compression mainly focused on the DCT (discrete cosine transform), but the recent researchmore focus on wavelet transform technique.Wavelet analysis is a method of analyzing the time-dimension of signal. Ithas a character of multi-resolution analysis, and also has the ability to detail local characters of a signal in both time and frequency. After a primitive image is transformed in wavelet, almost all of its energy is concentrated in the low frequency coefficients, minority is scattered in the high frequency coefficients. So how to organize the wavelet coefficient effectively is pivotal to the performance of image coding based on wavelet transform.In this paper, following works have been done:(1) After having realized the 5/3 and 9/7 wavelet lifting scheme, we made a deep study on EZW, SPIHT and EBCOT , which are the three representative wavelet coding algorithm. EBCOT which is the JPEG2000 standards core makes use of the correlation of neighbors in the same sub-band to organize context for high order adaptive arithmetic coding. While both EZW and SPIHT utilize the correlation of different sub-band to construct zerotrees, which is a variation of the classical coding.(2) We recompressed the SPIHT serial codes with Huffman and Arithmetic coding separately, error protection ability was also discussed at last.(3) Regions of interest coding with EBCOT and SPIHT arithmetic techniques were realized ,and their performance wre compared.(4) In th end, a parallel piplined architecture that performs the forward and inverse 5/3 discrete wavelet transform (DWT) is proposed by using a...
Keywords/Search Tags:Wavelet lifting scheme, EZW, SPIHT, EBCOT, Arithmetic coding, ROI coding, VLSI
PDF Full Text Request
Related items