Development of Poor Visibility Assessment Method in Winter Using Images Taken by On-Board Video Camera

抄録

In winter Hokkaido, poor visibility occurs due to snowstorm. It is required to develop a smart system to monitor visibility on the road to minimize traffic accidents. Thus, we create an enable system to assess visibility widely on the road using the on-board video cameras. The weighted intensity of the power spectrum (WIPS) was applied to identify poor visibility from images recorded by the on-board video cameras. First, we conducted the experiment to show driver’s subjective visibility assessment values regarding poor visibility conditions compared with the values of WIPS. Next, we investigated the changes of the values of WIPS under adverse weather conditions could reflect poor visibility along the road. As a result, values of driver’s subjective visibility assessments were close to the values of WIPS. The changes of the values of WIPS indicated changes of visibility widely on the road caused by meteorological conditions and road environments.