A Proposed Analysis and Digital Image Processing For Indication of Tailings Dam Failure

Abstract

Due to the recent accidents with ore tailings dams in Brazil, many researchers have been interested in alternatives to minimize the impact of such disasters. This paper proposes a method for analysis of surveillance images related to Córrego do Feijão mine dam, located in the city of Brumadinho, Minas Gerais, Brazil. The method uses digital image processing and computer vision algorithms to obtain an indication of the moment of its rupture. The results found show that the technique is relevant to the identification of dam rupture. The methodology in this paper can be used to develop autonomous monitoring solutions for the purpose of failure warning, which might provide time for people to take refuge in higher places before being reached by the slurry wave.