摘要
图像去噪是图像分割等操作的关键预处理环节,是从图像处理过度到图像分析和理解的重要手段之一,是图像处理中研究的重点内容。现有典型的一些图像去噪方法大多采用平滑滤波机理,在保护图像边缘信息等方面存在一定的局限性.本文在对现有典型的图像去噪方法总结的基础上,提出了一种图像去噪的新方法——利用支持向量机(SVM)技术构建图像去噪滤波器,并在实现算法和实验仿真方面做了初步探讨。
Image denoising is the key preprocessing for image segmentation, which plays an important role in image analyzing and understanding. The traditional approaches of image denoising are mostly on the basis of smooth filtering, and they are unable to protect edge image information effectively. In this paper, several typical approaches of image denoising are addressed briefly, then a new approach of image denoising based on Support Vector Machine (SVM) is presented, and the implementation of construction of image denoising filter and the experimental emulation with LIBSVM are discussed as well.
出处
《集宁师专学报》
2004年第4期17-21,共5页
Journal of Jining Teachers College
关键词
支持向量机
支持向量回归网络
图像去噪
Support Vector Machine(SVM)
Support Vector Regression Network
Image denoising
