An Estimation Method of Fragmentation in Blast Muckpiles using Local and Global Modules of Deep Learning Based on 3D Point Clouds

This paper proposes a method to estimate the particle (rock) size distribution of muckpile using Deep Learning based on 3D shape information given by photogrammetry. Optimization blasting is crucial increase productivity mining operations. However, since internal structure ground (bedrock) usually unknown, it difficult set appropriate parameters, amounts explosives, location and timing. In order solve problem, research has been carried out design analyze procedures measuring after blast. Ordinary works focus developing accurate measurement methods for analysis. We aim accuracy combining photogrammetry data. The model generated Structure from Motion (SfM), which reconstruction that can generate point clouds target object multi-view images. muckpiles estimated Learning. proposed network consists “Local Module” learns local rock “Global entire muckpile. From experiments, be said fragmentation actual effectively method.