基于自蛇模型的图像放大算法及实现  

Image Zooming Algorithm and Implementation based on Self-Snake Model

作　　者：任小斌 王佛荣 张善卿[2] 

机构地区：[1]河南中原高速公路股份有限公司郑开分公司,河南郑州410016 [2]杭州电子科技大学计算机图像研究所,浙江杭州310018

出　　处：《安阳师范学院学报》2015年第2期28-31,共4页Journal of Anyang Normal University

基　　金：浙江省自然科学基金项目<基于结构张量的纹理图像分割与修复>(项目编号:LY14F020043)

摘　　要：图像分辨率是决定图像质量的关键因素之一。目前有许多图像放大算法,但是,大多算法放大后的图像普遍性存在的问题是边缘模糊化和边缘锯齿化。为了克服这个缺点,可以把其当成是一种噪声,采用了滤波去噪的进行处理。本文采用的自蛇模型是一个非常有效的去噪模型,它在去噪和保持图像边缘方面,有非常优异的效果。因此,本文尝试运用自蛇模型处理图像放大时的边缘方片效应,并增加了校正的过程以提高复原后的图像质量。最后,对该方法进行了编程实验,对比并验证了该方法与传统方法相比的优越性。Image resolution ratio is one of the key factors that determine the image quality. There are currently many image zooming algorithms,although there are problems referred to blurred edges and jagged edges after the images have been magnified in most of the algorithms. In order to solve the problem which can be seen as image noise,filtering is adopted to deal with denoising. In this article,the self-snake model adopted is a very efficient de-noising model,it has outstanding results in de-noising and image edge preservation.Therefore,this article intends to apply the edge effects of self-snake model in the process of magnifying the images,along with added process of modification to enhance the restored image quality. The programming experiment has been carried out in the final part,tested and verified the favorable performance compared to traditional approaches.

关 键 词：图像放大 边缘方片效应 自蛇模型 去噪 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]