基于同类组和正则化的多帧超分辨率技术-计算机应用技术专业论文.docxVIP

华 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract Super resolution construction is a technology that produces a high resolution image from a set of low resolution images of the same scene, and one of the main branches of image enhancement. However, there are still some problems from the existing algorithms, as imaging models are assumed to be the white Gaussian noise, and they can hardly deal with the other types of additional noise, such as impulse noise. Furthermore, edge preserving performance is a particularly important measure. To solve these problems, we proposed two main improvements in this paper. Firstly, to apply to the impulse noise model, we adopt the peer group technology from the field image denoising. By considering the reliability of the pixels, weighted each pixel by the peer group way, and the weight represents the similarity between current pixel and its neighbors, then integrate the metrics into high-resolution image reconstruction. Secondly, the traditional BTV regularization method processes each pixel with the same weight, which doesnt consider the possibility of the noise, and not handle them properly. Our approach employs the reliability of each pixel to constrain the fidelity term of regularization formula, thus preserves more reliable features of low-resolution images, and attenuates the unreliable features, finally improves the quality of high-resolution images. Experiments shows that proposed algorithm achieves significant high-resolution result with edges simultaneously is robust to noise not only for white Gaussian noise, but also for the photon shot type outliers, such as impulse noise. Meanwhile, proposed algorithm is based on the similarity of spatial small range neighbor pixels, which is sample to implement and costs cheap in calculation. Keywords: Super Resolution, Peer Group, Regularization, Noise Robustness III III 目 录 摘 要 I Abstract II 1 绪论 1.1 研究背景和意义 (1) 1.2 国内外研究现状 (2) 1.3 论文研究内容及组织结构 (5) 2 超分辨率技术相关技术 2.1 超分辨率技术数学模型 (6) 2.2 系统噪