对星载合成孔径雷达(SAR)原始数据压缩算法的压缩性能评估是全面了解算法性能、合理选择算法方案的有效手段。该文模拟了以真实SAR图像为背景,引入点目标阵列的SAR回波数据;定义了综合评估指标;对分块自适应量化(BAQ)算法进行了不同分块方案、不同压缩比下的压缩实验;得到了一系列评估指标的计算结果。研究结果表明:BAQ算法对图像域的空间分辨率没有影响;不会引起图像的几何失真;对辐射分辨率影响很小;压缩引起的图像主瓣峰值失真不能够用统一的因子校正。同时该综合评价指标也可以推广到其他压缩算法评估中。

In order to know the performance of algorithm for space-borne SAR raw data compression and choose proper compression scheme, evaluation of compression algorithm is indispensable. By defining some of evaluation parameters, this paper utilizes actual ground simulation raw data to evaluate Block adaptive Quantization (BAQ) algorithm on different data separation scheme and compression ratio respectively. Numerical experiments show that BAQ algorithm neither affects spatial resolution nor generates geometric distortion. It has only a little effect on radiometric resolution. The distortion of image main-lobe peak can't be corrected uniformly. At the same time, this evaluation method can be used in evaluation of other algorithms for SAR raw data compression.

TN957.52