您的位置:首页  > 论文页面

基于显著性和视觉词典的图像分类算法研究

发表时间:2016-08-31  浏览量:1548  下载量:491
全部作者: 温静,赵雪
作者单位: 山西大学计算机与信息技术学院
摘 要: 针对盲图像无任何先验信息,必须依靠人的视觉感知特性指导分类,提出一种基于显著性和视觉词典的图像分类算法。首先对自然图像提取其显著性区域;其次对显著性区域进行加速鲁棒特征(speed-up robust feature,SURF)的提取;然后构造关于显著性的视觉词典;最后利用1-范数支持向量机(support vector machine,SVM)的稀疏性进行图像分类。实验结果显示,所提出的算法能够获得较好的分类效果。
关 键 词: 计算机应用;图像分类;Gist;加速鲁棒特征;显著性分割;多实例学习
Title: Image classification algorithm based on saliency detection and visual vocabulary
Author: WEN Jing, ZHAO Xue
Organization: School of Computer & Information Technology, Shanxi University, Taiyuan 030006, China
Abstract: In this paper, an image classification algorithm is proposed to deal with the blind image classification, which means the content of image is unknown, but the classification task could be guided by human visual perception. Firstly, the salient regions are extracted from natural images, then the speed-up robust feature (SURF) of the salient region is computed, after that, the visual vocabulary of the salient region is built up, the classification can be worked out by 1-norm support vector machine’ s sparse property. The experimental results show the efficiency and effectiveness of the presented approach.
Key words: computer application; image classification; Gist; speed-up robust feature; saliency segmentation; multiple instance learning
发表期数: 2016年8月第16期
引用格式: 温静,赵雪. 基于显著性和视觉词典的图像分类算法研究[J]. 中国科技论文在线精品论文,2016,9(16):1654-1660.
 
0 评论数 0
暂无评论
友情链接