摘要
将尺度不变特征变换(SIFT)算法应用到图像的特征点提取与匹配中,SIFT算法可在尺度空间寻找极值点,提取对图像焦距变化具有稳定性的特征点及其特征描述符。在采用SIFT算法提取图像的特征点及其特征描述符后,提出了一种特征点精匹配算法进行特征点的匹配,并通过仿真证明该算法具有很好的效果。
The scale invariant feature transform (SIFT) algorithm is applied to the image feature point extraction and matching. The SIFT algorithm can find out feature vectors in different scale spaces and extract stable image features and image description symbols with the invariance of focal length. An accurate matching algorithm for the feature points is proposed to match the feature points after the SIFT method was used to extract the feature points of an image and its description. The simulation results demonstrate the effectiveness of the algorithm.
出处
《现代电子技术》
2010年第23期116-118,共3页
Modern Electronics Technique
基金
国家自然科学基金项目(60672140)
关键词
多焦距
特征匹配
SIFT
特征提取
multi-focal length
feature matching
SIFT
feature extraction