摘要
针对尺度不变特征变换算法特征匹配时设置不变的欧氏距离比配准区间,该文提出根据图像自动调整配准参数的方法,以斐波那契数列项为长度分块,块内随机摆动探测,能够寻找到适应图像的距离比区间参数,同时运算量增加有限。实验表明,采用寻优方法在能够较快的找到最优配准区间参数。
Considering of the constraint of scale invariant feature transform algorithm when it set the parameter of the thresh of the distance ratio unchangeable,this paper proposed the method of adaptive parameter setting to suit for different images.We partition the region into n blocks by Fibonacci sequence and swing random to find the optimization value of the distance ratio,which don＇t cause too much calculation at the same time.The experiment shows that using the optimize method which proposed in this paper can find the optimization parameter in small iteration times.
出处
《杭州电子科技大学学报（自然科学版）》
2011年第5期175-178,共4页
Journal of Hangzhou Dianzi University：Natural Sciences
基金
黄山学院自然科学资助项目(xkj010)
关键词
尺度不变特征
斐波那契数列分块
随机摆动探测
图像自适应
scale invariant feature transform
Fibonacci sequence partition
random swing
adaptive parameter
