多图非局部去噪算法研究-电子与通信工程专业论文.docxVIP

万方数据 万方数据 摘要 摘 要 数字图像作为人类获取视觉信息的重要载体之一，其在采集和传输的过程中， 会不可避免地受到噪声的影响，导致数字图像的质量下降，阻碍人们从中获取所 需的信息。如何有效抑制噪声对图像质量的影响，为后续的图像处理提供更加真 实可靠的图像信息，一直以来都是数字图像处理领域研究的重点和热点问题之一。 对于传统的单帧图像去噪，其本质上是一个欠定性问题，通常是对噪声统计 特性进行一系列先验性假设进而获取近似的最优解。相比于单帧图像，针对同一 场景采集的多帧图像提供了更加丰富的空域信息，为图像去噪提供了新的研究思 路。本文针对同一场景的多帧图像去噪问题，从时域多帧图像存在更多空域信息 这一核心思想出发，分析和研究了多帧图像中空域信息的相关特性，提出了一种 多参考块的多图非局部均值去噪算法，并结合噪声统计特性分析，提出了一种自 适应权值的多图非局部均值去噪算法。 结合数字图像在变换域具有稀疏性的特点，本文从变换域单帧图像去噪算法 中的代表性算法——块匹配三维协同滤波去噪算法出发，研究多帧图像的应用场 景中，如何准确地实现相似块的搜索以及如何合理利用空时域相似块变换后的稀 疏性，提出了一种基于三维块匹配的变换域非局部去噪算法。 大量的仿真实验表明，本文所提出的使用自适应权值的多图非局部均值去噪 算法和基于三维块匹配的变换域多图非局部去噪算法在抑制噪声的同时，可以有 效地保留图像中的细节信息，相比于已有算法，在图像的主观质量上具有明显的 改善。 关键词：多帧图像去噪 空时域相似性 非局部均值 三维块匹配 多图非局部去噪算法研究 Abstract Abstract As one of the significant media which carry the visual information for human beings, digital images are tend to be influenced by noises during the acquisition or transmission, which results in a decline in the quality of images and prevents people from obtaining the visual information they need. How to effectively suppress the influence of noises on the image quality, to provide more realistic and reliable image information for subsequent image processing, has always been considered as one of the key areas and hot issues in the research field of digital image processing. Due to the ill-posed nature of the classical single image denoising problem, one must make some priori assumptions for the statistical properties of the noise to get an approximated optimal solution. However, compared to the single image, multi-images of the same scene captured in a short term can offer much more spatial information, which can be used for image denoising in a different way. In this work, the problem of image denoising by using multi-images of the same scene is addressed. Based on the analysis and study of the spatial correlation in multi-images in the same scene, a non-local means denoising algorithm with multi-reference blocks is proposed. Then with analyzing the statistical properties of the nois