基于区域Bhattacharyya相似度的SAR图像地物分类方法

Terrain classification of SAR images based on Bhattacharyya similarity between regions

传统的基于像素的合成孔径雷达(synthetic aperture radar,SAR)图像地物分类方法难以有效区分起伏变化大的地物。针对该问题,提出了一种基于区域Bhattacharyya相似度的SAR图像地物分类方法。方法首先利用适当的图像分割技术获取均匀的SAR图像区域。接着定义Bhattacharyya相似度来描述区域之间的统计相似程度,并推导了其对应Gamma分布的解析表达式。最后,以图像区域为分类单元,基于最大区域Bhattacharyya相似度准则实现SAR图像地物分类。利用实测SAR图像的地物分类结果表明,该方法性能优于经典的基于像素的最大似然分类方法和支持矢量机方法,且优于基于区域的最小距离法。

Terrains with large variance are difficult to be discriminated by the traditional pixel-based classifi- cation methods for synthetic aperture radar （SAR） images. Aiming at solving this problem, an algorithm for terrain classification of SAR images based on the Bhattacharyya similarity between regions is proposed. Firstly, a proper image segmentation technology is applied to obtain homogeneous regions of the SAR image. Then, the Bhattacharyya similarity measuring the statistical proximity between regions is defined, of which the analytical expression referring to the Gamma distribution is derived. Finally, with the previous image regions as classifica- tion elements, terrain classification is implemented by a criterion of maximizing the Bhattaeharyya similarity be- tween regions. The experimental results on the real SAR image validates the superiority of the proposed algo- rithm to the pixel-based maximum likelihood classification method, support vector machine classifier and the re- gion-based method of minimizing the distance between regions.