The Research And Application Of Embedded Coding In JPEG2000

JPEG2000 still image compression standard is the-state-of-the-art technique, which further improves the ability of current compression algorithm. Compared to JPEG, JPEG2000 possess many advantages such as lossy and lossless compression, region of interest, and support of low rate compression.JPEG2000 is consisted of 2 key algorithms discrete wavelet transform and embedded block coding with optimal truncation. Wavelete transform are better able to reduce redundancy between image pixels, providing coding framework of multi-resolution. Embedded block coding endow JPEG2000 with the feature of quality scalability, spatial scalability, progressive transmission, error resilience.This paper focus on the analysis of embedded block coding with optimal truncation and its process. This paper try to analyze the contribution to the image quality of independent block size, tile size and wavelet resolution level and give a general conclusion about how to choose the parameter. Due to the fact that under high rate compression most compressed bitstream would be discarded, during embedded block coding process a lot of time has been wasted on these coefficients. Therefore, this paper provide a method of predicting the truncation points of block coding to speed up block coding process.