论文部分内容阅读
分析了从空间遥感图象的自然背景中提取人造目标的复杂度和可行性，应用基于小波变换下多分辨率分解的特征融合技术，提出了一种从自然背景提取具有多边形外形人造目标的算法。对具典型意义的ＳＡＲ图象的人造目标检测，实验结果证明算法是有效的和有一定推广意义的。
The complexity and feasibility of extracting man-made objects from the natural background of spatial remote sensing images are analyzed. By using the feature fusion technique based on multi-resolution decomposition under wavelet transform, an algorithm of extracting artificial objects with polygon shape from natural backgrounds . Experiments on artificial targets with typical SAR images show that the algorithm is effective and has some generalizations.