毕业论文-基于中值滤波的图像椒盐噪声滤除算法的研究与应用.汇.docxVIP

本 科 生 毕 业 论 文（设计）题 目： 基于中值滤波的图像椒盐噪声 滤除算法的研究与应用 院 系： 信息科学与技术学院 专 业： 计算机科学与技术 学生姓名： 李翔 学 号： 指导教师： 罗笑南(教授) （职 称）二〇 年 月摘 要在现实生活中，将实际获得的图像进行数字化，并在其传输和处理过程中，不可避免的存在着一定的外部干扰和内部干扰，使用户所获得的数字图像被噪声污染，而这些噪声中椒盐噪声的表现更为明显。为此有大量学者针对椒盐噪声进行研究分析。中值滤波算法以其非线性的特点，在处理椒盐噪声上有显著的效果，因此许多学者针对中值滤波算法进行改进。典型的如加权中值滤波器、开关中值滤波器、自适应中值滤波器等。然而这些滤波器或在保护图像细节上能力不足，或去除噪声效果不佳，或对于高密度噪声无法处理，或过于复杂不便于硬件实现，存在着一定的缺陷。本文研究了几种典型的改进后的中值滤波算法，通过理论分析与实验仿真，比较其在去噪及保护图像细节各方面的优缺点。在此基础上，本文给出了一种基于噪声检测的模糊自适应中值滤波算法。在该算法中，通过设计窗口自适应的噪声检测机制，有效的区分了噪声与非噪声像素点，从而高效的保护了图像细节信息，并大量减少需要处理的像素点，降低算法运行时间。同时对于检测出的噪声点，则采用改进了的自适性中值滤除算法进行噪声滤除，在该滤波算法中加入了窗口自适应控制，对于高密度噪声也能有效的进行去噪处理。通过详细的理论分析后，本文基于matlab进行大量仿真实验，验证了这些新方法在噪声去除与细节保留上的有效性。关键词：图像去噪、椒盐噪声、噪声检测、自适应中值滤波、开关阈值滤波AbstractThe acquisition, recording and transmission of digital images through sensors or communication channels are often interfered by different types of noises, which may change the image. Impulse noise is one most common and important kind of noise. Impulse noise removal in image processing is an important pre-processing so that many researchers work on the restoration of images corrupted by impulse noise. Being the most popular nonlinear filter, the median filter is often used to remove impulse noise because of its good denoising power. In that case, many image-denoising filters are proposed based on the median filter, such as the weighted median filter, soft switching filter, adaptive median filter, etc.In this paper, a new image-denoising filter that is based on several advanced median filter is proposed. There is a two-phase scheme in this new algorithm. In the first phase, an impulse noise detection is used to identify pixels which are likely to be noise candidates. The noise detection has variable window size for removal of impulses, so that we can keep most of the signal content of the uncorrupted pixels, and time used for filtering can be reduced by a wide margin. In the second phase, the noise candidates will be filtered by the new filter. Based on the adapti