Wave runup video motion detection using the Radon Transform

Abstract

A new method of runup detection from video imagery is introduced and validated at an energetic dissipative beach. The instantaneous waterline is detected from uprush and backwash by using the Radon Transform (RT). The method is compared to conventional color contrast method from RGB images and LiDAR measurements. In our observations, the RT shows better detection skill even for adverse conditions, such as those present on flat dissipative swash zones where contrast is reduced. Because the RT is a proxy of deeper waterline (∼0.1 m) it is less sensitive to lack of contrast due to sand saturation. Moreover, since it is based on motion detection, it is less sensitive to changes in lighting conditions. Overall, the RT offers an attractive alternative for long term automated detection of the runup.