摘要
本文提出了一种特征点检测方法 ,通过引入局部特性因子和层次因子分别实现了自适应特征点检测及分层描述 ,不需输入参数 ,避免了复杂的参数试探过程。由于采用的主要是几何信息 ,算法简单 ,且与人类视觉系统有着很好的一致性。
A feature point detection algorithm is proposed in this paper.To realize adaptive point detection and hierarchical description,local attribute factors and hierarchy factors are introduced.Paramet er trial is avoided because no parameter is required to be inputted in this meth od.The geometric information adopted simplifies our algorithm and makes it more coherent with the human visual system.Experimental results are quite inspiring.
出处
《计算机工程与科学》
CSCD
2000年第6期16-18,共3页
Computer Engineering & Science
关键词
图像识别
数字曲线
特征点检测
图像匹配
feature point detection
polygonal app roximation
local attribute factor
hierarchy factor
scale-space analysi0
