摘要
在CCSDS建议的无损数据压缩的基本原理上 ,结合基于局域纹理特征的处理方法 ,给出数据输入输出在恒速信道条件下一种有效的面向空间应用的遥感图像无损压缩方案。该方案包括去相关处理、熵编码以及压缩缓存控制三部分。去相关操作是在局域纹理分析方法的基础上 ,通过提取两种有效的预测模型并自适应选择最佳的模型进行处理 ;然后采用快速有效的CCSDS熵编码器对最佳去相关模型的预测误差编码。在压缩方案中充分考虑了输入输出恒速信道条件下应用的特殊性 ,设计了相应的压缩码流控制方法 。
In the paper an effective image data lossless compression method is presented, which is mainly used for remote sensing image data transmitted via constant bit rate channel. The algorithm is composed of three parts, decorrelation preprocessor, adaptive entropy coder and output memory controlling. Decorrelation preprocessor comes from the theory of analyzing feature of local texture, the entropy coder is borrowed from lossless data compression recommendation given by the Consultative Committee for System (CCSDS), and the memory controlling is decided by actual condition of input and output channel. These problems of transmission error propagation and controlling of compressed data output buffer are all resolved effectively in the scheme.
出处
《中国空间科学技术》
EI
CSCD
北大核心
2003年第5期33-38,56,共7页
Chinese Space Science and Technology
