An approach for identifying classifiable regions of an image captured by autonomous robots in structural environments
详细信息 查看全文
- 作者：Andrew Wing Keung To ; Andrew.To-1@uts.edu.au" class="auth_mail" title="E-mail the corresponding author ; Gavin Paul Gavin.Paul-1@uts.edu.au" class="auth_mail" title="E-mail the corresponding author ; Dikai Liu Dikai.Liu@uts.edu.au" class="auth_mail" title="E-mail the corresponding author
- 关键词：Surface-type classification ; RGB-D ; Likelihood ; Texture features
- 刊名：Robotics and Computer-Integrated Manufacturing
- 出版年：2016
- 出版时间：February 2016
- 年：2016
- 卷：37
- 期：Complete
- 页码：90-102
- 全文大小：4229 K
文摘
- •
We have proposed an approach to identify regions in an image that can be accurately surface-type classified.
- •
Formulated a mathematical function to calculate the likelihood of accurate surface-type classification based on viewing distance and viewing angle.
- •
Examined a bridge surface inspection scenario where a robot is implemented to autonomously grit-blast surfaces as part of the preparation process of painting.
- •
Classified surface-types using Gray Level Co-occurrence Matrix features and naive Bayes classifier.