Reading Barcode signs to Support V21 Safety

Reading Barcode Signs to Support V21 Safety 3/20/2018 For this SBIR we will develop a machine vision system which uses onboard standard-resolution cameras in vehicles to identify and read barcode signs on the roadside. The proposed approach could provide a low-cost solution that make connected and automated vehicle (CAV) technology viable for rural areas with limited access to Information and Communications Technology (ICT). To be successful, vehicles must recognize and interpret content from signs, and reason about the extracted information to update navigation estimates, modify control behaviors, react to unsafe conditions, and trigger discrete events. Our approach builds upon previous R&D to provide a real-time architecture for detecting, tracking, enhancing, and recognizing barcode signs form dynamic video imagery. Detected barcodes are tracked to improve detection accuracy and provide resiliency to occlusions and environmental effects. The orientation of the sign is determined and the image is rectified and further enhanced for barcode decoding. Our Phase I program will result in thorough testing of the algorithm modules and an end-to-end feasibility demonstration on selected routes. In Phase II, a fully functional prototype system will be built and tested in a field test track.