Automatical Encoding of Button Products Based on Visual Recognition
Buttons is closely related to human life.The management of buttons in a button factory is very complex because there is a huge number of button types.In order to simplify the management complexity,this paper has proposed a method of multi-template matching to identify the buttons and automatically encode the buttons according to rules in the International Commodity Code so that button libarey can be searched using button pictures.Compared with the traditional template matching,the proposed method makes use of two sets of templates for rough classification,and then uses the texture combined with other features for fine classification,so that the unnecessary matching process can be reduced and the overall recognition rates can be improved to automatically encode button products.
Image segmentation Template matching Recognition of Buttons Feature extraction
Heng Yang QingXiang Wu PengFei Li Kai Lin Lei Hou
Key Laboratory of OptoElectronic Science and Technology for Medicine of Ministry of Education,College of Photonic and Electronic Engineering,Fujian Normal University,Fuzhou,China
国际会议
重庆
英文
399-403
2017-03-25（万方平台首次上网日期，不代表论文的发表时间）