摘要
传统的模板匹配算法(如NC和SSDA)虽然具有匹配精度高的优点,但由于其计算量大,使得匹配时间过长。本研究提出了一种新的基于投影与KMP简约算法的一维快速模板匹配算法,首先将图像进行降维处理,得到特征字符串,通过对特征字符串的匹配,实现粗匹配,可大幅度地减少图像匹配时间;然后利用NC匹配算法,对粗匹配的结果进行验证,以达到快速匹配的目的。通过实验证明了本算法的快速性和准确性。
Traditional template matching, such as the NC and the SSDA method have high matching precision, but the calculating are hard. They are time-consuming. The fast template matching algorithm based on the projection and the contracted KMP algorithmic has been proposed. Firstly, the feature strings is gotten by projecting the image followed by matching it to get the approximate matching result. Secondly, the NC method is used to validate the approximate matching result. The algorithm has been proven to be efficient.
出处
《青岛科技大学学报（自然科学版）》
CAS
2006年第2期176-178,共3页
Journal of Qingdao University of Science and Technology:Natural Science Edition
关键词
模板匹配
投影
特征字符串
KMP简约算法
template matching
projection
feature string
contracted KMP algorithm
