红外技术, 2018, 40 (7): 642, 网络出版: 2018-08-04
基于矢量量化与线性预测的干涉图分段无损压缩
Segmented Lossless Compression Method of Interferogram Based on Vector Quantization and Linear Prediction
无损压缩 干涉图 矢量量化 线性预测 压缩比 lossless compression interferogram Vector quantization(VQ) Linear prediction(LP) compression ratio(CR)
摘要
经计算证明了风云四号气象卫星上的干涉仪所采集的三维数据立方体在时间维上更容易取得高无损压缩比,于是决定以时间维上的单帧干涉数据为单位进行无损压缩的探索。结合干涉数据在时间维上的相关性特点,提出一种新的无损压缩算法。该算法根据干涉数据的各段特点,分别进行矢量量化和线性预测以此来进行压缩,采用MATLAB 对实际采集数据进行仿真试验后可得,新方法最终取得的无损压缩比比直接进行Huffman 编码所得的压缩比提高了10%~20%。
Abstract
The three-dimensional data cube collected by an interferometer on the FY-4 was analyzed, and we concluded that a high compression ratio can be more easily achieved in the time dimension; therefore, we studied the lossless compression on the single-frame interferogram of time dimension. A new lossless compression algorithm was proposed based on the interferogram characteristics. According to the characteristics, the interferogram was segmented, and vector quantization and linear prediction were respectively used for compression. Finally, the results simulated in Matlab show that a 10% to 20% higher compression ratio can be achieved by the new algorithm than an algorithm directly encoded by Huffman.
杨敏珠, 邹曜璞, 韩昌佩. 基于矢量量化与线性预测的干涉图分段无损压缩[J]. 红外技术, 2018, 40(7): 642. YANG Minzhu, ZOU Yaopu, HAN Changpei. Segmented Lossless Compression Method of Interferogram Based on Vector Quantization and Linear Prediction[J]. Infrared Technology, 2018, 40(7): 642.