Fumito Shinmura

In this paper, we propose a method of classifying road congestion status over wide road areas using satellite images. Using our method, we first extract the road areas from satellite images using a digital map. Next, we extract the front and rear of vehicles by detecting the edges on road areas and by eliminating several noises, and we compute the number of vehicles on each road section from the number of the front and rear of vehicles. Then, we compute the density of vehicles in each road section by dividing the number of vehicles by the area of the road section. Road sections having a high density are considered to be congested, thus, the congested status can be classified. Although there is currently a problem with the extraction error, our study shows the possibility of collecting traffic information directly from satellite images. © 2011, The Institute of Image Electronics Engineers of Japan. All rights reserved.