An intelligent framework for end?to?end rockfall detection

Rockfall detection is a crucial procedure in the field of geology, which helps to reduce associated risks. Currently, geologists identify rockfall events almost manually utilizing point cloud and imagery data obtained from different caption devices such as Terrestrial Laser Scanner (TLS) or digital cameras. Multitemporal comparison clouds with these techniques requires tedious visual inspection implies inaccuracies that depend on several factors human expertize sensibility sensors. This paper addresses this issue provides an intelligent framework for event any individual working intersection geology domain decision support systems. The development analysis presents major research challenges justifies exhaustive experimental analysis. In particular, we propose system utilizes multiple machine learning algorithms detect clusters data. Due extremely imbalanced nature problem, plethora state-of-the-art resampling accompanied by models feature selection procedures are being investigated. Various pipeline combinations have been examined benchmarked applying well-known metrics be incorporated into our system. Specifically, developed applied them analyze extracted TLS two distinct case studies, involving geological contexts: basaltic cliff Castellfollit de la Roca conglomerate Montserrat Massif, both located Spain. Our results indicate some above-mentioned pipelines can utilized incidents mountain walls, experimentally validated accuracy.