DEVELOPMENT OF A VISIBILITY FORECAST MODEL BASED ON A ROAD VISIBILITY INFORMATION SYSTEM AND METEOROLOGICAL DATA

Abstract

The study proposes a model that forecasts visibility in winter by using either multiple-regression analysis or the Kalman filter. We have been developing the Road Visibility Information System (RVIS), which calculates road visibility information as the weighted intensity of power spectra (WIPS) and the present-time road visibility index (RVI) from daytime road images recorded by multiple closed-circuit television (CCTV) cameras along the roads. The objective of this study is to develop the visibility forecast model based on 1-km-mesh meteorological data. We used data of the WIPS values and RVI ranks recorded by the RVIS and 1-km-mesh meteorological data recorded by the Japan Weather Association during the winter of 2009-2010 at a 35-km section of National Route 40 in Hokkaido, Japan. A multiple-regression model and the Kalman filter were employed to reveal the relationship between WIPS data from road images as a dependent variable and the meteorological data as independent variables. The Kalman filter can be regarded as the preferable of the two visibility forecast models examined in the study. Also, the 1-km-mesh meteorological data of air temperature, wind speed and snowfall were determined to be informative independent variables in the forecast models.