Wavelet-Based Model of Grain-Size Distribution Detection Applied to UAV Images along the Shimizu Coast

Abstract

 Continuous beach sediment mapping of the Shimizu coast in Japan has been obtained by a drone, collecting more than 600 images at a fixed height of 10 meters above the mean sea level along a 5 km stretch of the coast. For each image, the bidimensional wavelet transform was applied to determine the sediment size of grains with a diameter larger than 10 mm. Through the application of such a model, the continuous grain-size distribution of the Shimizu coast has been successfully estimated. The obtained grain-size distribution along the study coast shows that diameters tend to decrease in the direction of littoral transport. In addition, the wavelet method results reveal unforeseen and peculiar local trends in the grain-size distribution, which have not been detected by low-density traditional laboratory sieve analyses.