遥感图像数据实时无失真压缩的研究

讨论了对单波段遥感图像和多波段遥感图像都适用的一种实时无失真数据压缩的方法，该方法应用线性预测原理减少数据的冗余度并应用Huffman编码以实现最佳平均码长，文中讨论和比较了几种预测的方法，并选用了2－D空间预测，为方便实现实时的数据压缩，本文对Huffman编码作了必要的简化，对于不同编码技术的结果，文中也进行了讨论，本方法将软件仿真应用于LANDSAT TM，SPOT HRV，AIR SAR以及中科院上海技术物理研究所研制的航空成谱仪的8bit／像元遥感数据，平均压缩比为2，其中最高的为3．1，最低的为1．2。

STUDY ON REAL-TIME LOSSLESS DATA COMPRESSION FOR REMOTE SENSING IMAGES

The paper describes a real-time lossless data compression algorithm for both individual band and multi-spectral remote sensing images. This approach uses prediction to reduce the redundancy and classical Huffman coding to reduce the average code length. Several kinds of predictions were studied and the 2-D prediction was selected. Some simplifications were taken in the Huffman coding for real-time use. A comparison between the coding methods was discussed. The scheme was applied to the 8 bits/pixel, original images of LANDSAT TM, SPOT HRV, AIR SAR and imaging spectrometer by software emulation. An average compression ratio of 2, with the highest value of 3.1 and the lowest value of 1.2 for the radar images, was achieved.