一种基于近似有限Ridgelet变换的SAR图像分割方法

An Approximate Finite Ridgelet Transform Algorithm for SAR Image Segmentation

由Donoho等提出的有限正交Ridgelet变换成功应用于高噪声图像的边缘检测和分割,但由于有限Radon的“缠绕”现象使得在该方法图像的重构时产生边缘的“混叠”和“洞”,影响了边缘检测和图像分割的质量。论文结合Wedgelet变换提出了基于“自然直线”的近似有限Ridgelet变换从根本上克服了这些缺陷和解决了离散Radon变换的图像重构问题。最后将这一方法用于SAR图像分割,并取得了满意的结果。

The Finite Ridgelet Transform that was recently proposed by Donoho have the successful extraction of edges and object in high-noise image,but it has a drawback which is due to annoying“wraparound”effect to the Finite Radon Transform.As a result,that may also produce“alias”edges and“hole”in the reconstructed image.In this paper,we will present a new finite redgelet transform which based on the“natural line”and Wedgelet Transform could overcome these weakness and solves the problem of Inverse Discrete Radon Transform.At last,we segment SAR image by this algorithm and get the satisfied segmentation.