Real-time Rotationally Invariant Features for Environmental Feature Detection by Mobile Robots Sensor Networks

Abstract

We introduce a mobile and/ or remote sensor framework for computationally fast rotationally invariant feature detection. The sensor and computational system is small enough to be carried by a mobile robot platform with a relatively low power requirement allowing the system to be deployed without the need for frequent recharges of the batteries. The rotationally invariant Haar-like features are introduced and evaluated both at feature level and in classifiers. Other invariant approaches such as moment based approaches do not offer the same discriminatory power as the Haar-like rotationally invariant features to detect complex objects such as hands and faces.