文献类型：期刊文献

中文题名：一种基于序贯相似性检测算法(SSDA)的加速算法

英文题名：A Speed-up Algorithm Based on SSDA

作者：沈慧玲[1];戴本祁[1]

机构：[1]华东理工大学信息学院,上海200237

年份：2006

卷号：21

期号：4

起止页码：60

中文期刊名：光电技术应用

外文期刊名：Electro-Optic Technology Application

语种：中文

中文关键词：模板匹配;SSDA;序贯相似性检测算法;加速算法

外文关键词：model matching; SSDA; speed-up algorithm

摘要：序贯相似性检测算法(SSDA)是针对传统模板匹配算法提出的一种高效的图像匹配算法,在该算法中引进了阈值,即模板与其覆盖下的搜索子图的像素差值的上限值.提出的基于SSDA的加速算法,是在此基础上,引入了模板与其覆盖下的搜索子图的像素差值的最低门限值,如果该最低门限值超过了阈值,则可以跳过该搜索子图.实验证明:该算法比传统的SSDA算法在速度上有所提高.
The sequential similarity detection algorithm （SSDA） is studied to improve the conventional image matching,which is an efficient image matching algorithm. A threshold is given in the algorithm, which means the upper bound of distance between the template and the corresponding window in the image. A speed-up algorithm is proposed. The lower bound of distance between the template and the corresponding window is calculated and if the value exceeds the threshold, this window can be skipped. Experiments show that the proposed algorithm has a improvement in the speed compared to SSDA.