doi:10.16186/j.cnki.1673-9787.2020.4.15

Received:2019/12/16

Revised:2020/02/25

Published:2020/07/15

A fast recognition algorithm for improved template matching

LI Changyou1, FU Huiqi2, CHEN Guoxi1

1.School of Mechanical and Power Engineering， Henan Polytechnic University ，Jiaozuo  454000 ， Henan， China;2.Department of Electrical Engineering ，Jiaozuo Technician College ，Jiaozuo  454000 ， Henan， China

Abstract:In order to solve the problem of mismatch caused by noise， and to improve the speed and accuracy of recognition algorithm，a fast recognition algorithm with improved template matching was proposed. A novel low- pass filter was used to perform noise reduction on the image. Then，an improved weighted Hamming Distance was used to perform similarity detection on the image to reduce the complexity of the algorithm. Finally，the threshold and homotopy skeleton algorithms were used to speed up the recognition speed.By experimenting with noise images in different backgrounds,the proposed algorithm had a good matching effect in the target recognition process， and had good robustness to noise,and improved the efficiency of the algorithm. Thus the effectiveness of the algorithm was proved.

Key words:template matching;low pass filter;weighted Hamming Distance;Homotopy skeletons

改进模板匹配的快速识别算法_李长有.pdf