Robust motion filtering as an enabler to video stabilization for a tele-operated mobile robot

Abstract

An increasing number of inspection and hazardous environment tasks use mobile robotic vehicles manually tele-operated via a live video feed from an on-board camera. The resulting video imagery frequently suffers from vibration artefacts compromising the accuracy and security of operation in addition to the viable duration for human tele-operation. Here we aim to automatically remove these unwanted visual effects using a novel real-time video stabilization approach. Prior work for hand-held and vehicle mounted cameras is ill-suited to the high-frequency, large magnitude (10-15% of image size) vibration encountered on the short wheelbase, non-suspended robotic platforms typically deployed for such tasks. Without prior knowledge of the robot ego-motion (or vibration characteristics) we develop a novel four stage filtering approach to identify robust Local Motion Vectors (LMV) for Global Motion Vector (GMV) estimation in successive video frames whilst preserving the required real-time responsiveness for tele-operation. Experimental results over a range of tele-operation scenarios show that the method provides both significant qualitative visual improvement and a quantitative reduction in measurable video image displacement (caused by vibration).