摘要
通过对灰度相关法特征点匹配算法的理论研究和实验分析,提出了一种能够克服图像尺度变化的特征点匹配算法。该算法主要根据图像特征点间欧氏距离的关系,结合传统的特征点灰度相关法和特征点的梯度相关法进行精确匹配。实验证明,该算法容易理解,易于实现,匹配结果较精确,误匹配点较少。
A feature point matching algorithm which can overcome the image scale change is proposed by theoretical research and experimental analysis of feature points matching algorithm of gray correlation method. It is mainly based on the relationship of Euclidean distance of image feature points com- bined with the traditional gray correlation method of feature points and the gradient correlation method of feature points to precise matching. Experiments prove that it is easy to understand ,easy to implement, have more accurate matching results and less false matching points.
出处
《电视技术》
北大核心
2014年第3期4-6,47,共4页
Video Engineering
基金
陕西省科技研究发展计划工业攻关项目(2013K07-35)
西安市科技计划项目(CXY1343(4)
CX1258⑤
CX1258⑥)
陕西省教育厅科研计划项目(12JK0508
12JK0535)
西安市碑林区应用技术研发项目(GX1209
GX1311
GX1310)
关键词
欧氏距离
相关法
特征点梯度
尺度变化
euclidean distance
correlation method
gradient feature points
scale change
