针对遥感图象随机性强，相关性差以及遥感图象数据采集速率高的特点，提出了根据图象局部区域纹理特征的复杂程度区别对待的自适应按块量化压缩编码的思想，并在此基础上构成多模式自适应量化压缩模型。计算机仿真结果表明，在压缩比为4的条件下，上述模型的压缩速度与恢复图象精度均明显优于国际静图压缩标准JPEG所得的结果。

In account for the strong randomness .weak local correlation and extra-high data acqusition rate of the remote sensing image.a multi-mode adaptive quantization image compression method suitable for remote sensing image is presented in this paper. The basic idea of the method is to code each image subareament using different quantization level based on the local texture complexity. This method is easy to implement and has better restored image quality compared with the JPEG’s