Proposal of collecting lighting situation of roads at night for recommendation of safety walking route using smartphone

This paper proposes a method for collecting road lighting situation at night for recommending safety walking route. In recent years, there is a growing demand for public route information service. Among them, the lighting situation, i.e. brightness of a road at night is important information for public safety. Such information is collected and visualized by some municipalities, but the coverage of the survey is not sufficient due to measurement equipment costs and labor costs. Other researches focus on lighting situation of street lamps, but these researches ignores the influence of other lights to the road illumination such as vending machines. In order to estimate lighting situation of a road, this paper proposes three attributes, which are calculated from illuminance data collected from a road. Illuminance data is collected using a light sensor of a smartphone. In order to evaluate the effectiveness of these attributes, k-NN and Naive Bayes classifiers are built from actual road illumination data.