Share Email Print

Proceedings Paper

Stereovision-based vegetable row recognition algorithm for agricultural vehicles
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

It is a new field for applying stereovision into guidance system. The objective was to find out a better correlation method and develop an algorithm for detecting vegetable rows for field robot. Several Area correlation methods were compared for obtaining disparities, such as sum of absolute differences and Mahalanobis Distance. A method to eliminate error matched results was also studied, which was comparing of minimum extremum and the second minimum extremum. The 3D data of fields were calculated based on the disparity images and they were matched with a trapezium model to detect vegetable rows. It shows that stereovision could obtain the landforms of fields, especially that of vegetable rows. In future, a real time tilt angle sensor might be added for more reliable 3D data.

Paper Details

Date Published: 12 October 2006
PDF: 10 pages
Proc. SPIE 6382, Two- and Three-Dimensional Methods for Inspection and Metrology IV, 63820O (12 October 2006); doi: 10.1117/12.686013
Show Author Affiliations
Fangming Zhang, Zhejiang Univ. (China)
Tong Sun, Zhejiang Univ. (China)
Yibin Ying, Zhejiang Univ. (China)

Published in SPIE Proceedings Vol. 6382:
Two- and Three-Dimensional Methods for Inspection and Metrology IV
Peisen S. Huang, Editor(s)

© SPIE. Terms of Use
Back to Top