一种基于区域分割的SIFT图像特征提取算法  

A SIFT image feature extraction algorithm based on region segmentation

作　　者：陶铭洋 郭阶添[1] 韩立新[1] 陈肖[1] 孙周宝[1] 王敏[1] 

机构地区：[1]河海大学计算机与信息学院,南京210000

出　　处：《信息技术》2013年第10期126-130,共5页Information Technology

摘　　要：提出一种基于区域分割的SIFT图像特征提取算法。首先采用改进后的SIFT算法对图像进行计算,同时采用快速分割算法对原始图像进行分割,对照分割后的区域分别取每个区域最显著的几个SIFT特征,最后采用局部降维算法,将高维特征降低到可以接受的低维度特征。实验表明,该方法的运行时间短,对特征提取的位置界定准。This paper proposes a SIFt image feature extraction algorithm based on region segmentation. Firstly, it introduces an improved SIFT algorithm to extract the features, at the same time, it imposes a fast segmentation algorithm to split the original image into areas, and then takes the most significantly first few SIFT features respectively in each region, at last, it uses a dimensional reduction algorithm to get desire low dimension. The experimentai results show that the operating time of the method is quite short, and it has a great definition of the location of the features.

关 键 词：自动标注 特征提取 区域分割 SIFT 

分 类 号：TP393.41[自动化与计算机技术—计算机应用技术]