基于块的比特分配的无损图像压缩算法

摘要

为了获得最佳的无损图像压缩效果，提出一种基于块的比特分配技术（ block bit allocation ，BBA）的图像无损压缩算法，并采用三种数字图像模式：CT、MR和DR来评价该算法。实验结果显示，此算法对比标准Lempel－Ziv和Huffman压缩算法能将CT图像的压缩比平均增加11％到19％，在MR图像中，平均增加12％到28％。对于DR图像，此方法比Huffman算法差9％。但是，压缩和解压缩速度会更快并且可用于在线医学图像压缩。%To achieve an optimal lossless image compression , an improved method using block bit allocation technology was proposed .Images from three digital modalities of computed tomography ( CT) ,magnetic resonance imaging ( MR) and digital radiography ( DR) were used to evaluate the technique .Preliminary results showed that this method could achieve a 11%to 19% increase in the compression ratio for CT images and a 12% to 28%increase in the compression ratio for MR images as compared to standard Lempel-Ziv and Huffman coding methods.For DR images,this method performed about 9% worse than Huffman coding .However,the compression and decompression of this method was much faster and could be implemented on -line for teleradiology application .