论文部分内容阅读
以大图像块或整个图像为处理单元的图像编码算法需要大量的内存来缓存图像,且编码过程中也会消耗大量内存,这种直接分块算法往往带来方块效应,影响图像的恢复质量。提出了以重叠块为单位的提升小波变换的方法,重叠分块可减小编码器对大块内存的需求,同时还可去除分块引入的方块效应。在变换中提出了多级并行分解方法,提高了分解效率。在对重叠块提升小波变换后的子带进行了统计分析,采用了DPCM与SPIHT相结合的方法。对直接分块、重叠分块、不分块算法进行了对比实验。结果表明,经重叠分块算法压缩的遥感图像具有较高的恢复质量。
The image coding algorithm using large image block or entire image as processing unit needs a large amount of memory to cache the image, and also consumes a large amount of memory in the coding process. This direct block algorithm often brings the square effect and affects the image restoration quality. A method of lifting wavelet transform based on overlapping blocks is proposed. Overlapping blocks can reduce the encoder’s requirement for large blocks of memory, and can also remove the block effect introduced by blocks. In the transformation, a multi-level parallel decomposition method is proposed to improve the decomposition efficiency. The subbands after wavelet transform of overlapped block are statistically analyzed, and a combination of DPCM and SPIHT is adopted. On the direct block, overlapping block, no block algorithm for comparative experiments. The results show that the remote sensing image compressed by the overlapped block algorithm has high recovery quality.