DocumentCode :
638900
Title :
An identification and classification method for circular object based on rotating image template matching
Author :
Jianke Chen ; Chao Hu ; Xiaoying Yuan ; Zhongqing Feng ; Houxiang Miao
Author_Institution :
Taiyuan Univ. of Sci. & Technol., Taiyuan, China
fYear :
2013
fDate :
4-7 Aug. 2013
Firstpage :
1338
Lastpage :
1343
Abstract :
This paper describes a method for automatically identifying and classifying various circular objects by image processing and machine vision technology. Firstly, the object images are acquired by an industrial camera, and the object range is determined, which is preprocessed by binary and noise filtering; Secondly, a region growing method and a circle-touching are used to find out the center of circular object; Finally, circular template matching method is used to extract the similarity feature of different circular objects, so as to determine the type of target circular object. The experiments prove the validity of this method.
Keywords :
computer vision; feature extraction; filtering theory; image classification; image matching; object detection; binary filtering; circle-touching; circular object classification method; circular object identification method; circular template matching method; image processing technology; industrial camera; machine vision technology; noise filtering; object range; region growing method; rotating image template matching; similarity feature extraction; Assembly; Feature extraction; Noise; Object recognition; Real-time systems; Wheels; circular object; circular template matching; identification and classification;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Mechatronics and Automation (ICMA), 2013 IEEE International Conference on
Conference_Location :
Takamatsu
Print_ISBN :
978-1-4673-5557-5
Type :
conf
DOI :
10.1109/ICMA.2013.6618107
Filename :
6618107
Link To Document :
بازگشت