液晶与显示, 2014, 29 (5): 751, 网络出版: 2014-08-18   

运动模糊图像点扩散函数的频谱估计法

Motionblurred image point spread function estimation method based on spectral analysis
作者单位
1 中国科学院 长春光学精密机械与物理研究所,吉林 长春 130033
2 中国科学院大学,北京 100049
摘要
在深入研究无噪声情况下的匀速直线运动模糊图像的点扩散函数特性,提出一种基于频域的识别运动模糊图像模糊方向和模糊长度的算法。分析了对模糊图像频谱图出现平行暗条纹的原因,从理论和实验说明暗条纹的方向与模糊方向垂直的特性,并采用RADON变换法检测这些暗条纹,给出精确的数值估算方法。采用SOBEL算子对模糊图像进行一阶微分运算,求其自相关鉴别曲线,并估算出其模糊长度。仿真分析结果表明,该算法可以准确地识别图像运动模糊参数,其模糊角度绝对误差不超过2°,模糊长度绝对误差不超过1个像素。
Abstract
This paper studied the characteristics of uniform linear motion blurred image of the point spread function in noiseless case, and proposed an algorithm which identify the direction of motionblurred image blur and blur length in frequency domain. Firstly, the causes of parallel dark stripes appeared in blurred image spectrogram were analyzed, and the characteristics of the direction of dark stripes perpendicular to fuzzy orientation were explained both in the theoretical and experimental methods. Then the RADON transform method was used to detect these dark stripes, for giving precise numerical estimation method. The SOBEL operator was used as firstorder differential operators to the fuzzy image, to seek its autocorrelation curve, and estimates its blur length. The simulation analysis shows that the algorithm can accurately identify the image motion blur parameters, the absolute error of angle and length does not exceed 2 ° and 1 pixel, respectively.

贤光, 颜昌翔, 张新洁. 运动模糊图像点扩散函数的频谱估计法[J]. 液晶与显示, 2014, 29(5): 751. XIAN Guang, YAN Changxiang, ZHANG Xinjie. Motionblurred image point spread function estimation method based on spectral analysis[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(5): 751.

本文已被 5 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!