基于霍夫变换的图像运动模糊角度识别法的改进  

Improvement of motion blurred direction identification based on Hough transform

作　　者：黄琦[1] 张国基[1] 唐向京[1] 

机构地区：[1]华南理工大学数学科学学院,广州510640

出　　处：《计算机应用》2008年第1期211-213,共3页journal of Computer Applications

摘　　要：运动模糊图像的复原主要依赖于对运动模糊参数的估计。针对R.Lokhande等人提出的基于霍夫变换的运动模糊角度估计法进行改进,通过增加预处理步骤,并利用霍夫变换得到更多合适直线,然后取加权均值得到最后估计角度值。实验表明,改进方法能得到比原方法更精确的角度值,并且更加稳定。Restoration of the motion blur images is highly dependent on estimation of motion blur parameters. The motion blurred direction identification based on Hough Transform, which was proposed by R. Lokhande et al., was improved. The improvement adopted more pretreatments, found more appropriate lines during the process of Hough Transform, and then gained the weighted means of their corresponding direction as the final result. The experiments show that, the improved algorithm has better precision and is more robust.

关 键 词：运动模糊参数估计 图像复原 霍夫变换 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]