论文部分内容阅读
EZBC算法综合利用了子带内和子带间系数的相关性,把零树/零块结构和基于上下文编码的优点有机结合在一起,获得了比SPHIT算法更好的压缩性能,比EBCOT更高的压缩效率。但是EZBC算法编码中的两个排序链表需要很大且非固定的存储空间,这使得EZBC算法的硬件实现非常困难。在EZBC算法的基础上提出了一种易于硬件实现、低存储量、高压缩性能的内嵌零块图像编码算法。该算法利用比特平面节点重要性状态表和上下文查找表来完成整个编码过程和形成上下文。实验结果表明,所提出的算法具有与EZBC算法基本相同的高压缩性能,但所需存储空间约为EZBC算法的四分之一,所以该算法更易于硬件实现。
EZBC algorithm makes full use of the correlation between sub-band and sub-band coefficients, combines the zero-tree / zero-block structure with the advantages of context-based coding, and obtains better compression performance than SPHIT algorithm, which is higher than EBCOT Compression efficiency. However, the two sorted lists in EZBC algorithm coding need large and non-fixed storage space, which makes the hardware implementation of EZBC algorithm very difficult. Based on the EZBC algorithm, an embedded zero-block image coding algorithm that is easy to implement with hardware, low memory and high compression performance is proposed. The algorithm uses the bit plane node importance state table and context lookup table to complete the entire coding process and form the context. The experimental results show that the proposed algorithm has the same high compression performance as the EZBC algorithm, but the required storage space is about one-fourth that of the EZBC algorithm, so the algorithm is easier to implement in hardware.