An Automatic Visibility Measurement System Based on Video Cameras

Principal Investigator:

Taek Kwon, Professor, UMD-Electrical Engineering

Project Summary:

In 1995, the Federal Highway Administration (FHWA) published study results on a comprehensive list of environmental sensor systems related to safe traffic operation. As a part of this study, the research team analyzed the technologies available for low-visibility detection, and particularly noted that video detection techniques have not been used in the field, although it has been frequently used for manual verification of on-site low-visibility conditions. They pointed out that since video-based sensors are increasingly being utilized in the transportation field for vehicle and incident detection, it can provide a cost effective solution if the same hardware could also be used for low-visibility detection. The purpose of this research is to further the initial findings of the FHWA study mentioned above by developing a prototype that can be used in the field in order to automatically quantify visibility from video images. The system includes a warning function that makes an alarm call to the maintenance engineers if visibility falls below a certain level that requires a warning. This project will deliver a working prototype that can be directly used in the field without modification.

Sponsor:

Project Details: