›› 2018, Vol. 31 ›› Issue (1): 54-.

• 论文 • 上一篇    下一篇

基于模糊核优化的图像盲去模糊算法

 尤佳旗   

  1. 上海理工大学 机械工程学院
  • 出版日期:2018-01-15 发布日期:2018-01-11
  • 作者简介:尤佳旗(1992-),男,硕士研究生。研究方向:图像处理。
  • 基金资助:

    国家自然科学基金 (51475310);教育部高等学校博士学科点专项科研基金(20133120120005)

Blind Image Deblurring Based on Optimized Fuzzy Kernel

YOU Jiaqi   

  1. School of Mechanical Engineering, University of Shanghai for Science and Technology
  • Online:2018-01-15 Published:2018-01-11

摘要:

针对图像的模糊算法优化问题,首先选取高斯分布拟合自然图像的分布特性,利用双边滤波器从模糊图像中提取出清晰的图像边缘。针对降噪进行参数设置,在初步估计出模糊核之后,对模糊核进行正规化修正工作。最后在图像复原阶段,利用优化的凸函数拟合自然图像分布,并利用快速傅里叶变换提高算法计算速度。实验结果,表明该模糊核优化算法与现有的其他算法相比,复原后的图像具有更好的视觉效果,且计算时间减少约20%。

关键词: 高斯分布, 滤波器, 模糊核, 傅里叶变换

Abstract:

In view of optimizing image fuzzy algorithm,distribution characteristics of natural images is fitted by selecting gaussian distribution firstly. The bilateral filter is used to extract image edges from blurred image. Setting parameter for noise reduction, after preliminary estimate the fuzzy kernel, fuzzy kernel revision is normalized. Finally, in image restoration stage, optimized convex function is used to fit distribution of natural images and fast Fourier transform algorithm are used to improve calculation speed. The experimental results show that compared with other existing algorithms, this image fuzzy optimization algorithm has better visual effect and significantly reducing computing time about 20% after recovery.

Key words: gaussian distribution;filter;fuzzy kernel;the Fourier transform

中图分类号: 

  • TP18