计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (30): 188-191.

• 图形、图像、模式识别 • 上一篇    下一篇

运动模糊图像点扩散函数参数识别的新方法

范海菊,刘艳芳,冯乃勤   

  1. 河南师范大学 计算机与信息技术学院,河南 新乡 453007
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-10-21 发布日期:2011-10-21

New parameter identification method of point spread function from motion blurred image

FAN Haiju,LIU Yanfang,FENG Naiqin   

  1. School of Computer and Information Science,Henan Normal University,Xinxiang,Henan 453007,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-10-21 Published:2011-10-21

摘要: 为了估计任意大小运动模糊图像点扩散函数,提出了真实方向、测量方向和列宽行高比的角度变换公式。在此基础上利用Radon变换最大值曲线检测峰值角度作为候选方向,剔除候选方向中Radon变换未出现规律性波谷的干扰方向后为待变换方向,再进行角度变换即为识别方向。取该方向上Radon变换向量,利用相邻波谷位置差剔除奇异值求均值的方法检测暗条纹间距,通过曲线拟合鉴定模糊尺度。仿真结果表明算法实时性好、精度高、抗干扰能力强。

关键词: 点扩散函数, 列宽行高比, 模糊尺度, radon变换, 曲线拟合

Abstract: To estimate the point spread function of any size motion blurred image,perspective transformation formula between measuring direction and true direction is built about column width row height ratio.In accordance with the above views,to begin with detecting peak angular of maximum based on radon transformation curve to serve as candidate orientation,removing interference direction that does not appear regularity trough,then the final identification direction is the remaining angular after perspective transformation.Radon transform curve under blurred direction can be generated.The pitch of black strips is detected by the mean value of adjacent wave trough location differences in curve after removing singular value.Then blurred scale is identified by curve fitting.Simulation result shows that this algorithm not only has good real-time calculation and high precision,but also has strong immunity.

Key words: point spread function, column width row height ratio, blurred scale, radon transformation, curve fitting