Vision modeling optimization of freely placed and moving objects

DOI

10.2991/icmmse-17.2017.67How to use a DOI?

Keywords

Robotic task, Machine vision, Visual feature matching, Taguchi methods

Abstract

For the intelligent operations using robots and vision sensors, it is a seemingly simple but in fact complex problem to design an efficient and accurate visual feature matching program. To determine the optimum identification pattern of freely placed and moving objects, the orthogonal experimental approach was employed. Firstly, a complex part that had eight visual features was selected as the experimental parts. Secondly, the experiment scheme was planned as a L12(211) orthogonal array. Thirdly, the robotic handling experiments were performed. Finally, the experimental data were analyzed and the results indicated that different identification patterns of the shape and alignment would lead to different success rates, and the maximal deviation of success rates reached 60%. Furthermore, according to the variance analysis the optimum identification pattern was determined

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hu Fuwen
AU  - Li Li
AU  - He Yunhua
PY  - 2017/04
DA  - 2017/04
TI  - Vision modeling optimization of freely placed and moving objects
BT  - Proceedings of the Second International Conference on Mechanics, Materials and Structural Engineering (ICMMSE 2017)
PB  - Atlantis Press
SP  - 403
EP  - 406
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmmse-17.2017.67
DO  - 10.2991/icmmse-17.2017.67
ID  - Fuwen2017/04
ER  -