影像科学与光化学 ›› 2018, Vol. 36 ›› Issue (2): 187-191.DOI: 10.7517/j.issn.1674-0475.2018.02.008

• 论文 • 上一篇    下一篇

再现图像细节抑制散斑噪声技术研究

吴育民1, 段海燕2, 文永富1, 程灏波1, 王谭1   

  1. 1. 北京理工大学 深圳研究院, 广东 深圳 518057;
    2. 北京空间机电研究所, 北京 100083
  • 收稿日期:2017-07-13 修回日期:2017-09-20 出版日期:2018-03-15 发布日期:2018-03-15
  • 通讯作者: 程灏波
  • 基金资助:
    深圳市科技创新计划资助项目(JCYJ20170412112640428,JCYJ20170224172022809)

Study on Denoising Technology to Reproduction Image Detail

WU Yumin1, DUAN Haiyan2, WEN Yongfu1, CHENG Haobo1, WANG Tan1   

  1. 1. Shenzhen Research Institute, Beijing Institute of Technology, Shenzhen 518057, Guangdong, P. R. China;
    2. Beijing Institute of Space Mechanics and Electricity, Beijing 100083, P. R. China
  • Received:2017-07-13 Revised:2017-09-20 Online:2018-03-15 Published:2018-03-15

摘要: 为了消除数字全息再现像产生的散斑噪声,一种基于边缘检测各向异性扩散方程的去噪算法被提出。在各向异性扩散方程(PM)基础上,通过将边缘检测算子结合梯度算子引导扩散过程,边缘检测算子对噪声具有低敏感度,可以更好地区分边缘信息,达到数字全息再现像去噪目的。通过去噪效果图对比和与中值滤波、均值滤波、小波变换滤波算法峰值信噪比(PNSR)的比较,表明本文提出的算法具有更好的去噪效果,在去除噪声的同时有效保护了图像的细节信息。

关键词: 数字全息, 散斑去噪, 各项异性扩散, 边缘检测

Abstract: In order to eliminate speckle noise in digital holographic reconstructed image, a denoising method based on edge detection anisotropic diffusion equation (PM) is proposed. The diffusion process is guided by the edge detection operator instead of the gradient operator. The edge detection operator is less sensitive to noise, and can better recognize the edge information of reconstructed image. Compared to median filtering, mean filtering, wavelet transform filtering algorithm of peak signal-to-noise ratio (PNSR), the experiments show that the proposed algorithm has better denoising effect, and can effectively protect the details of the image in the process of denoising.

Key words: digital holography, speckle denoising, anisotropic diffusion, edge detection