摘要
针对目前尺度不变的图像特征点提取算法计算量较大,算法较复杂的问题,提出一种简化的SIFT图像特征点提取算法。此算法通过改变金字塔尺度空间的结构实现对SIFT特征点提取过程的简化,通过改变特征点描述子的结构实现对特征向量计算的简化,从而在保证算法鲁棒性的同时减少了计算量并增强了实时性。实验证明了该算法的有效性。
The scale-invariant image feature point detecting methods are always complex and need large computation. In order to solve the problem, this paper proposed a feature point detecting method which was a simplification of the SIFT method. The method changed the pyramid frame in image scale space to simplify the SIFT feature point detecting process and changed the descriptor configuration to simplify the eigenvector computation. It could ensure the performance and decrease the computation at the same time. The experimental results have proved its validity.
出处
《计算机应用研究》
CSCD
北大核心
2008年第7期2213-2215,2222,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(60675028)
关键词
特征点提取
图像匹配
尺度不变特征变换算法
feature point detection
image matching
SIFT（scale invariant feature transform） method
