基于分形理论的SAR图像分割

SAR(合成孔径雷达)图像包含有相干斑噪声,传统方法不能很好对其进行分割,文章将SAR图像的特点和分形理论相结合,提出了一种新的SAR图像分割方法。该方法首先对原始SAR图像每个像元为中心取两种不同窗口,计算在该两种窗口下基于区域自选的分形维数并求均值,将其作为分形纹理特征。然后根据SAR图像噪声在小波域中的分布特点对SAR图像进行滤波,最后以SAR图像分形纹理特征和滤波后的灰度组成特征向量对SAR图像进行分割。实验结果分析表明,该方法是一种有效的SAR图像分割方法。

A New Method of SAR Image Segmentation Based on the Fractal Theory

For speckle consisted in SAR image,good segmentation result can't be gotten with traditional methods.In the paper,a new method of SAR Image segmentation is proposed with the combination of the features of SAR image and fractal theory.In the method,firstly,with two different kinds of windows around each image pixel,the fractal dimensions are distilled based on the adaptively region choosing method,then the average value of the two fractal dimensions is given to the image pixel as the fractal texture feature.Second,SAR image is filtered according to the features of its dis-tribution in the wavelet field,at last,SAR image is segmented according to the vector composed of the fractal texture features and gray level of filtered SAR image.The experiment results show that the new method is efficient in SAR im-age segmentation.