Region Of Interest Coding Based On Jpeg2000

With the increasing use of multimedia technologies, image comcression requires higher performance as well as new features. To address this need in the specific a rea of still image encoding, a new standard is currently being developed, the JPEG 2000. Different from the JPEG standard, JPEG2000 bases on discrete wavelet transf orm and the newest embedded block coding with optimized truncation algorithm--E BCOT, so it not only provides superior compression performance over JPEG, but al so provides a set of new features. ROI coding is one of the features. ROI coding could not only satisfy the requirement that the important image information coule b een restored in high-quality even nondestructive in the condition of low bit rate;bu t also combinate with the image to provide a very good solution to the bandwidth bottlenecks, storage space problems in communication. Due to the obvious dvantage s, ROI coding becomes a research hotspot in recent years.Based on Image Compression Standard JPEG2000, this paper discussed and studie d about EBCOT image compression algorithm, region-of-interest encoding techniques a nd realization aspects.include:(1) Studied the EBCOT algorithm principle and key technologies used, including indep endent embedded block coding techniques, rate-distortion optimization, and flexible bit-stream organization. At the same time, analyzed and researched the basic prin ciples and technology realization of the Region-of-interest encoding.(2) Based on a a comprehensive analysis the main principles and mechanisms for cont ent of the EBCOT algorithm, according to drawbacks of Tier-1 of EBCOT coding algorithm:computation is to big and and Processing time is to long,it proposrd An optimization program, that is, parallel processing and coding channel parallel p rocessing. The experimental results show that it improves the encoding speed and enhanced the real-time.(3) Based on a comprehensive analysis of the Maxshift Method and the Scaling-based Method, it proposed a novel Region-Of-Interest (ROI) coding method, which cou Id adaptively shift the rate-distortion slopes of the ROI code-blocks. The proposed method could avoid the low efficiency caused by increasing bit-planes.Experimen tal results demonstrate that the proposed method enables the flexible adjustment of image quality between ROI and background, can support arbitrarily shaped ROI coding without coding the shape, and can reconstruct ROI with higherquality more quickly.