Automatic Detection of Snowfall Intensity Using CCD Sensor

摘要

Due to the global warming and climate change, economic damage caused by heavy snowfall is drastically increased in recent years. Sudden heavy snowfall in the road area causes traffic disruption and congestion. To reduce the damage, appropriate countermeasure should be executed based on the real-time snowfall intensity information for each road segment. In this paper, we present a newly developed approach for the automatic detection of snowfall intensity using CCD sensor in the roadway. The proposed approach starts from determining the analyzing zone in the image space. The zone is called MLZ (MotionLess Zone). MLZ is selected on the condition that it does not contain lane marking, safety zone marking, and moving cars. Next, snow streaks, which trespasses the MLZ, are extracted with proposed algorithm. The proposed algorithm consists of edge detection, extracting parallel line segments, and excluding noise. Also, we propose the concept of SII(Snowfall Intensity Index) which is the number of snow streaks trespassing the MLZ during one minutes. The effectiveness of proposed approach is evaluated qualitatively comparing with real imagery and SII.