Abstract
Image template matching is applied to various fields such as the industrial image applications. This paper proposes a new method for image template matching that is based on the position of the pixel with the median gray-level in a local image area. The position of the pixel that has the just intermediate gray-level in the order of all gray-levels of pixels in a local image area is used for the matching. The matching is performed between the templates that have the positioning addresses of the pixels with the median gray-levels and an objective gray-scaled image. The experimental results show that the sharp peaks of the best solutions and the less noises around the peaks were obtained by the proposed method in comparison with the conventional methods. The processing time for the matching was 40% shorter than the normalized correlation matching.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
