A Fast Template Matching Using Adaptive Window Skipping Method Considering Position of Reference Template

作者：Toyama Fubito^*; Shoji Kenji; Miyamichi Juichi

来源：Electronics and Communications in Japan, 2011, 94(12): 45-56.

DOI：10.1002/ecj.10343

• template matching
• adaptive window skipping method (AWS)
• SSDA
• lower bound of distance

The Adaptive Window Skipping method (AWS) is known as a fast template matching algorithm. This method can reduce the computational burden by using similarity between subtemplates. AWS guarantees the same accuracy as an exhaustive search with template matching. In AWS preprocessing, a reference subtemplate is selected from the set of subtemplates. The position of the reference subtemplate affects the computational burden. But this position has not been taken into account in AWS. In this paper we propose a new method that takes into account the position of the reference subtemplate. Experimental results show that the proposed method is faster than AWS.

• 出版日期2011-12