Deep Learning Based Granularity Detection Network for Mine Dump Materials

The granularity distribution of mine dump materials has received extensive attention as an essential research basis for stability and land reclamation. Image analysis is widely used the fastest most efficient method to obtain materials. This article proposes a deep learning-based approach detection identification material, conglomerate, clay. Firstly, Conglomerate Clay Dataset (CCD) proposed study dump. A typical area selected field sampling, sampled conglomerate clay photographed labeled. In addition, this keypoint-based algorithm detection. considers scale variation in orthophoto images adopts center point avoid difficulty localization. On basis, dense convolution introduced feature extraction reduce computational redundancy conduct more efficiently. Finally, corresponding distributions are obtained by geometric calculation results. validated on dataset CCD, experiments demonstrate effectiveness its application material.