基于模糊推理的图像放大算法  

Image Zoom Method Based on Fuzzy Reasoning

作　　者：干宗良[1] 齐丽娜[1] 

机构地区：[1]南京邮电大学江苏省图像处理与图像通信重点实验室,南京210003

出　　处：《中国电子科学研究院学报》2010年第5期498-502,共5页Journal of China Academy of Electronics and Information Technology

基　　金：国家自然科学基金项目(60802021)资助课题

摘　　要：图像放大技术在许多图像应用系统中具有重要作用。图像的主要特性之一是具有模糊特性,将模糊处理方法引入图像放大算法中,提出一种基于模糊推理的图像放大算法。该算法首先建立了基于模糊推理的图像模型,假设图像相邻的像素点是一种特定的模糊推理规则,并且在局部窗口之内,这种模糊推理规则保持不变。在基于模糊推理的图像放大算法中,根据基于模糊推理的图像模型预测内插像素点的数值,并且这种模糊推理规则可以根据局部像素点的训练得到。实验结果表明,所提出的基于模糊推理的图像放大方法能够有效地提高图像的主客观质量。Image zoom technique plays an important role in many image application systems.Since fuzzy feature is one of the main image characteristics,a novel image zoom method based on fuzzy reasoning is proposed.Firstly,the image model based on fuzzy reasoning is set up.In the proposed image model,adjacent image pixels are considered as a kind of specific fuzzy inference rules and the fuzzy rules remain constant within the local window.Then,the proposed image zoom method is given,in which the interpolation image pixels are estimated by the fuzzy reasoning,and the fuzzy inference rules are calculated based on the training of local pixels.Experiments show that the proposed approach can improve image quality effectively.

关 键 词：图像放大 信号内插 模糊推理 模糊规则训练 

分 类 号：TN911.73[电子电信—通信与信息系统]