计算机应用研究2013，Vol.30Issue(3)：924-925,941,3.DOI:10.3969/j.issn.1001-3695.2013.03.073
一种基于SIFT算法的图像镜像变换识别方法
Method based on SIFT for recognition of flip images
摘要
Abstract
The image feature extracted by SIFT algorithm does not match the mirror transform of image itself very well. Aiming at this problem,this paper modified the SIFT algorithm, and proposed the FI-SIFT algorithm. The FI-SIFT algorithm firstly used the gradient of Y-axis to determine whether it was a mirror transform, if it was, mapped the descriptor to polar coordinate. After the reassembling the polar coordinates, the new descriptor could be obtained in the Cartesian coordinates by reverse transform. Then matched the critical points by Euler distance. The experimental results show that the image features obtained by the FI-SIFT algorithm match the mirror transformation of the image much better. The average matching accuracy rises from 0.6 to 0.9.关键词
尺度不变特征变换算法/镜像不变—尺度不变特征变换算法/图像特征/镜像变换/匹配精度Key words
SIFT algorithm/ FI-SIFT algorithm/ image feature/ flip transform/ matching accuracy分类
信息技术与安全科学引用本文复制引用
及莹,陆月明..一种基于SIFT算法的图像镜像变换识别方法[J].计算机应用研究,2013,30(3):924-925,941,3.基金项目
国家"863"计划资助项目(2011AA01A205) （2011AA01A205）
