| With rapid mergers of computer, communications, and entertainment industries, we can expect a trend of growing heterogeneity (in channel bandwidth, receiver capacity, etc.) for future digital image and video coding applications. The traditional multi-bitstream approach to the heterogeneity issue is very constrained and inefficient. Scalable coding, allowing partial decoding at a variety of resolution and quality levels from a single compressed codestream, is widely considered as a promising technology for efficient signal representation and transmission in a heterogeneous environment. However, although several scalable coding algorithms have been proposed in the literature and the international standards over the past decade, these former methods can only accommodate relatively limited decoding flexibility at a significant cost of overall coding performance.; In this thesis a family of coding systems built upon a new subband/wavelet bitplane coding framework, embedded zero block coding (EZBC), is developed for a variety of digital image and video scalable coding applications. The two powerful embedded wavelet coding techniques—set partitioning and context modeling—are efficiently combined in the proposed wavelet bitplane coding algorithm. We demonstrated through extensive experiments that our new coding systems performed comparable or better than the state of art in compression while possessing some attractive features such as low computational complexity and high coding scalability.; We exploit the natural multi-resolution and multi-precision structure in a subband/wavelet bitplane coding system to accommodate many attractive scalability functionalities. A hierarchically structured codestream is created to facilitate efficient retrieval of the desired image/video resolution from the compressed file. Encoding and decoding are allowed to be effectively separated so that the decoding quality will not be limited by the quality and bitrate constraints determined at encoding times. As is exhibited in our experiments, such a high coding scalability in the proposed coding system is achieved without an significant cost at complexity and performance commonly experienced in most scalable coding systems. |
