For this report, this project’s research team set out to develop a system to analyze videos from the second Strategic Highway Research Program Naturalistic Driving Study dataset and automatically produce annotations and descriptors for events, behavior, and driving scenarios related to transportation safety. The project had four objectives: characterize high-level driver behavior, such as eating or using a cellphone; classify the environment outside the vehicle, such as the position of roadway objects, work zones, and intersections; understand interactions and dependencies between drivers and the surrounding environment, such as looking at a billboard or a passing vehicle; and demonstrate how the video analytics techniques used in this study can help human factors researchers address research questions in novel ways.

Recommended citation: Federal Highway Administration, Using Video Analytics to Automatically Annotate Driver Behavior and Context in Naturalistic Driving Data (Washington, DC: 2024) https://doi.org/10.21949/1521750