黄欢, 聂庆文. 基于前向与后向扩散模型的变分去噪[J]. 云南大学学报(自然科学版), 2016, 38(3): 369-375. doi: 10.7540/j.ynu.20150248
引用本文: 黄欢, 聂庆文. 基于前向与后向扩散模型的变分去噪[J]. 云南大学学报(自然科学版), 2016, 38(3): 369-375. doi: 10.7540/j.ynu.20150248
HUANG Huan, NIE Qing-wen. Image denoise based on FAB diffusion model and ROF[J]. Journal of Yunnan University: Natural Sciences Edition, 2016, 38(3): 369-375. DOI: 10.7540/j.ynu.20150248
Citation: HUANG Huan, NIE Qing-wen. Image denoise based on FAB diffusion model and ROF[J]. Journal of Yunnan University: Natural Sciences Edition, 2016, 38(3): 369-375. DOI: 10.7540/j.ynu.20150248

基于前向与后向扩散模型的变分去噪

Image denoise based on FAB diffusion model and ROF

  • 摘要: 基于偏微分方程的算法是近年来图像增强中广泛运用的去噪技术.对于图像去噪,单一地采用ROF算法无法有效地保留图像的细节信息,也会产生局部的阶梯效应.针对这个问题,提出结合全变分和前向与后向扩散模型的算法,不仅有效地锐化边缘效果,而且控制了阶梯效应的产生.最后实验结果表明两种方法的结合对于图像有更高的峰值信噪比,改善了单一方法对图像去噪的局限性.

     

    Abstract: Algorithm based on partial differential equation is widely used in image enhancement denoising technology in recent years.For image denoising,using PDE algorithm in a single can't effectively preserve image detail information,and can produce the steps of the local effect.Aiming at this problem,this paper puts forward the algorithm combinated total variation and the forward and backward diffusion model not only sharpen edges effect effectively,but also control the generation of the staircase effect.The simulation results show that the combination of the two methods for image has higher peak signal to noise ratio,and improves the limitations of single method for image denoising.

     

/

返回文章
返回