Machine Vision System to Support Vehicle to Infrastructure (V21) Safety Applications

To keep drivers and passengers of vehicles safe, especially in rural areas with limited access to Information and Communications Technology (ICT), the DOT and the commercial automotive industry are keenly interested in machine vision based V2I and CAV technology to perform functions such as navigating assistance in areas where GPS or detailed maps are unavailable, caution and warning systems capable of recognizing unsafe conditions, and timely notification of changing road conditions. To address this interest, IAI has developed RB3S system in Phase I, with several demonstrated capabilities including: 1) Robust barcode road sign locating and recognition algorithm based on machine learning; 2) Integrated system with real-time recognition capability; 3) Dynamic road condition retrieval; and 4) testing on one-lane suburb roads. In Phase II, IAI will refine and enhance the RB3S system for practical use in highway environments. Some of the proposed Phase II efforts include: 1) Detailed requirement analysis in real highway environments; 2) System ruggedization and miniaturization; 3) Improvement of the barcode recognition method under challenging image conditions with longer distance, and 4) Robust test in real multi-lane highway conditions in both suburb and urban environments.