【作者】 刘振宇；

【导师】 顾玲嘉；

【作者基本信息】 吉林大学 ， 工程硕士（专业学位）， 2018， 硕士

【摘要】 图像是人类获取信息的重要途径,图像质量的好坏关系到人类获取信息的多少。去噪的最终目的是改善给定的图像,解决实际图像由于噪声干扰而导致图像质量下降的问题。通过去噪技术可以有效地提高图像质量,增大信噪比,更好的体现原来图像所携带的信息,作为一种重要的预处理手段,人们对图像去噪算法进行了广泛的研究。在现有的去噪算法中,有的去噪算法在低维信号图像处理中取得较好的效果,却不适用于高维信号图像处理;或者去噪效果较好,却丢失部分图像边缘信息,或者致力于研究检测图像边缘信息,保留图像细节。如何在抵制噪音和保留细节上找到一个较好的平衡点,成为近年来研究的重点。图像去噪过程就是为了减少图像中的噪声,提高图像质量。图像去噪算法要求在去除噪声的同时,还要尽可能的保留图像的边缘细节结构。传统的去噪方法难以做到两者兼顾,而本文所研究的偏微分图像去噪算法是一个新兴交叉学科分支,在理论分析和去噪效果上要优于传统图像去噪算法,能够有效地保持去除噪声和保留边缘细节结构之间的平衡,并为构造新的去噪模型提供理论支持。本文主要介绍了图像去噪的基础知识,如基本概念、基本理论和基础算法等,研究并分析了偏微分方程表达式,推导了偏微分方程去噪算法的两个经典模型PM模型和TV模型,并给出数值实验结果。更多还原

【Abstract】 Images are important sources of information for human beings,the quality of images is related to how much information can be received.The ultimate goal of denoising is to improve the given image and solve the problem that the image quality is degraded due to noise interference.Through denoising technology can effectively improve the image quality,increase the signal-to-noise ratio,better reflect the original image of the information carried,as an important pretreatment means,people to the image denoising algorithm has been widely studied.In the existing denoising algorithm,some denoising algorithms can achieve better results in low dimensional signal image processing,but they do not apply to high dimensional signal image processing,or the denoising effect is good,but some image edge information is lost,or the image edge information is studied,and the details are preserved.How to find a better balance in resisting noise and keeping details has become the focus of research in recent years.The image denoising process is to reduce the noise in the image,as well as improve the image quality.The image denoising algorithms require removing the noise while preserving the edge details of the image as much as possible.The partial differential image denoising algorithm studied in this paper is an emerging cross-disciplinary branch which is superior to the traditional image denoising algorithm in the theoretical analysis and denoising effect,and can effectively maintain the balance between removing noise and preserving the edge detail structure,as well as provide theoretical support for constructing new denoising models.This article introduces the basic knowledge of image denoising in detail,such as basic concepts,basic theories and basic algorithms.The expressions of partial differential equations are studied and analyzed.Two classic model---PM models and TV models are deduced,and the numerical experimental results are given.更多还原