一种基于混合整型变换和3D-SPIHT的高光谱图像嵌入式无损压缩方法  

Hyperspectral Image Lossless Compression Algorithm Based on Hybrid Transform and 3D-SPHIT

作　　者：柴焱[1] 计文平[1] 沈兰荪[1] 

机构地区：[1]北京工业大学信号与信息处理研究室,北京100022

出　　处：《电子学报》2007年第9期1770-1773,共4页Acta Electronica Sinica

基　　金：国家自然科学基金(No.60472036;90304001);北京市自然科学基金(No.4052007);北京市教委(No.KM200410005022);北京市科技新星计划(No.2005B08)

摘　　要：利用高光谱图像空间、谱间相关性不同的特点,本文提出了一种基于混合整型变换和三维分层树集合划分算法(3D-SPIHT)的高光谱图像无损压缩方法,首先将波段进行分组,针对每一分组,采用不同的整型变换技术去相关,然后对变换系数进行系数重组,采用高效的基于小波系数特点的3D-SPIHT编码方法得到嵌入式码流,具有一定的抗误码性能.实验结果表明,该方法能够有效的去除高光谱图像的空间和谱间相关性,与现有的国际标准JPEG-LS、JPEG2000和基于三维离散余弦变换(3D-DCT)或三维离散小波变换(3D-DWT)的编码方法相比,压缩后的平均比特率均有明显降低.Using the spatial and spectral correlation within the hyperspectral images,a lossless compression algorithm is presented. Band grouping is used to divide images into different groups.Different hybrid integer Iransform based on spatial/spectral correlation is taken for each group and to achieve progressive bit-stream, coefficients are reorganized to obtain the same structure as wavelet Iransformation. Transformation coefficients are then coded by bit-plane-based 3D-SPIHT coding algorithm. Experimental resuits show that the proposed method can remove the spatial and spectral redundancy more efficiently than JPEG-LS, JPEG2000 and the algorithms based on integer 3D-DCT or 3D-DWT. With band grouping and progressive bit-stream, the proposed method can achieve better error resilient performance.

关 键 词：高光谱图像 无损压缩 整型变换 系数重组 三维分层树集合划分算法 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]