多波段遥感图像的快速分形编码

Fast Fractal Coding of Multispectral Remote Sensing Images

分形图像编码用动态迭代过程的参数来表示相应的静态图像数据 ,可以突破熵压缩编码的理论界限 ,因而得到广泛的关注 .将分形图像编码应用于多波段遥感图像压缩中 ,根据多波段遥感图像的特点 ,对基于四叉树分割的快速分形编码方法进行改进 ,令各个波段图像共享同一四叉树分割方式 ,同时 ,利用多波段遥感图像的谱间相关性 ,缩小仿射变换的搜索匹配空间 ,以此来提高压缩比和编码速度 ,实验表明 ,本方法能显著提高基于四叉树分割的分形编码方法的性能 ,取得了令人满意的结果 .

Fracatal image coding which can emplog parameters of the dynamic processes of iteration to represent the static image data and break through the theoretical bound of entropy coding has now received wide attention.In this paper,the fractal image coding is employed to compress the multispectral remote sensing images.Making use of the properties of the multispectral remote sensing images,some improvememts are made to the fast fractal image coding method based on the quadtree partitioning.The compression ratio and the coding speed were both improved by sharing the same quadtree partitioning of the range blocks with other different image bands and,in the same time,reducing the size of the searching and matching space of the affine transform by utilizing the spectral correlation of the multispectral remote sensing images.The experimental results show that the performance of the fast fractal coding method based on quadtree partitioning was significantly improved by incorporating our scheme.Satisfactory experimental results are obtained.