摘要
利用Unigraphics NX软件进行数字化纸张的产品建模,主要针对在纤维成纸过程中基于点阵图形的数字化纸张信息的模拟,从理论角度形成数字化的纸张图像。进而根据模拟纸张信息的特性,采用基于图像像素值的模板匹配、寻找特征点匹配的ORB算法以及改进后的ORB算法对纸张信息进行识别效果的对比。实验结果表明,在保持识别准确度的前提下,改进后的ORB算法耗时较短,具有较强的实时性,为纸张的功能数字化信息的识别提供了理论基础。
Product modeling of digital paper was conducted through Unigraphics NX,mainly focusing on the simulation of digital paper information based on dot matrix graphics during the paper forming process,and the digitized paper images were obtained theoretically.The image identification results for paper information using template matching based on image pixel values,ORB and improved ORB were compared.The experimental results showed that the improved ORB algorithm was time-consuming and had strong real-time performance under the premise of keeping the matching accuracy,which provided a theoretical basis for the identification of digital functions of paper.
作者
张开生
韦逸野
ZHANG Kaisheng;WEI Yiye(School of Electrical and Information Engineering,Shaanxi University of Science and Technology,Xi′an,Shaanxi Province,710021)
出处
《中国造纸学报》
CAS
CSCD
北大核心
2018年第1期61-66,共6页
Transactions of China Pulp and Paper
基金
2017年陕西省科技计划项目(编号:2017GY-063
项目:特种纸生产过程中内嵌信息的数字化表征与识别研究)
关键词
数字化纸张
建模
图像识别
ORB算法
digital paper
modeling
image identification
ORB algorithm