基于GM模型和Hilbert空间填充曲线的图象压缩方法

本文提出了一种新的基于灰色模型 GM(1,1)和 Hilbert空间填充曲线的灰度图象压缩方法 .我们利用 Hilbert空间填充曲线来选取模型化序列中的象素数据 ,并据此使用 GM(1,1)模型化图象象素 .由于 Hilbert曲线是一种能够最好地保持空间点的局部邻接性的扫描曲线 ,因此基于 Hilbert曲线的这种数据选取方法能改进图象的压缩比并且显著地降低编码误差 .实验结果表明 ,本文给出的方法能获得误差小于 4%且压缩比小于 10 %.

A Image Compression Method Based on GM and Hilbert Space-Filling Curve

In this paper, a novel image compression approach for the compression of grey scale images based on GM(1,1) and Hilbert space-filling curve is proposed. We use Hilbert space-filling curve to scan image to obtain the modeling sequence, and then model the pixels of the sequence by GM(1,1). As Hilbert curve is the scanning curve that is best to preserve the local proximity of the points in the domain space, the data selection method can improve the compression rate and reduce the error rate significantly. The experimental results show that the compression rate is less than 10% under 4% error.