Project summary:
This research explored the formulation and implementation of an automated computer-vision and machine-learning-based system for estimating the occupancy of passenger vehicles in high-occupancy vehicles and high-occupancy toll (HOV/HOT) lanes. The research employed a multimodal approach involving near-infrared images and high-resolution color video images in conjunction with strong maximum-margin-based classifiers such as support vector machines. The researchers attempted to maximize the information that can be extracted from these two types of images by computing different features, then built classifiers for each type of feature that were compared to determine the best feature for each imaging method. Based on the performance of the classifiers, the researchers critiqued the efficacy of the individual approaches, as the costs involved are significantly different.