Research For Region-of-Interest Coding Algorithm Based On JPEG2000

Computer is widely applied to process image and other multimedia. How to compress image effectively for saving resources becomes a research hot spot at present. Among international standards designed for still image compression, JPEG2000 is the newest standard. Different from the JPEG standard, JPEG2000 bases on discrete wavelet transform and the newest embedded block coding with optimized truncation algorithm, so it not only provides superior compression performance over JPEG, but also provides a set of new features. ROI coding is one of the features. The ROI coding technology utilizes different compression rate to different area in an image. It can solve the contradiction between the image quality and the compression rate.This paper summarizes the general information of JPEG2000 still image compression standard and discusses each part of the JPEG2000 coder in detail according to the JPEG2000 encoding order. The main research work and contributions of the thesis are summarized as follows:1. An improved image compression algorithm is proposed, which is based on image extraction and JPEG2000. Prior to being compressed by JPEG2000 coder, the original image is extracted firstly, and this pretreatment can reduce the size of the image which is entered into JPEG2000 system. The experimental results show that this improved algorithm is simple and effective, and it can not only reduce computing complexity, but also keep the quality of the reconstructed image.2. A modified ROI coding algorithm is proposed. According to this algorithm, all of the wavelet coefficients in the lowpass subband are processed firstly, and then all of the ROI coefficients in the rest highpass subbands are processed, finally all of the BG coefficients in the highpass subbands are processed. The experimental results show that this modified algorithm can get the same effect of ROI coding, but having no need for coefficient displacement and dynamic extension spread.