The rear-view mirror is unhelpful when an overtaking car is in the blind quadrants (blind spot). In this contribution we describe the software implementation of an algorithm to monitor vehicle overtaking processes. This algorithm detects the vehicle to the rear, and discriminates whether it is approaching or not, and if approaching, it alerts us of its presence. The proposed system is based on the Reichardt correlator model [1]. The approach presented uses the saliency of motion features in a competition scheme to filter noise patterns. In this way features corresponding to rigid body motion self-emerge from the background. Real overtaking sequences have been to develop this monitoring system.
No takes yet. Share an insight, caveat, or question.
Mota et al. (2004) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: