This paper presents a novel real-time obstacle detection system based on stereoscopic vision. The system was designed for an Unmanned Ground Vehicle (UGV) operating in an unstructured environment. The system is based on the Scale Invariant Feature Transform (SIFT) Algorithm. It permits to detect, track and reference obstacles with respect to a fixed inertial reference system. Experimental results are shown to demonstrate the validity of the approach.
|Titolo:||Stereo Vision Obstacle Detection Based on Scale Invariant Feature Transform Algorithms|
|Anno del prodotto:||2007|
|Appare nelle tipologie:||4.1 Contributo in Atti di convegno|