一种运动模糊医学图像去模糊算法分析

A Deblurring Algorithm Analysis of Medical Motion Blur Images

针对运动模糊医学图像去模糊需求设计一种修复算法,分析运动模糊医学图像的基本原理,引入汉宁窗,减小频谱泄露,降低频谱幅值,以减小噪声,引入radon变换,估算运动模糊角度;分析自相关图估算模糊长度算法的不足,引入逐行一阶微分图像,估算运动模糊长度;重构点扩散函数,实现图像去模糊。与当前主流算法就运动模糊角度和运动模糊长度的估算结果进行分析和比较,结果表明文章算法图像去模糊效果更好。

The repair algorithm is designed for the de-blurring demand of medical motion-blurred images. The basic principle of motion blur medical image is analyzed, and the Hanning Window is introduced to reduce the spectrum leakage, to reduce the spectrum amplitude, to reduce the noise. The Radon transform is introduced to estimate the motion blur Angle. This paper analyzes the defect of the algorithm for estimating the fuzzy length of autocorrelation graph,and introduces the first-order differential to estimate the fuzzy image length of motion. The Point Spread Function is reconstructed to realize the image de-blurring. The estimation results of motion fuzzy direction and motion fuzzy length are analyzed and compared with the current mainstream algorithm.The result shows that the image de-blurring effect of this algorithm is better.