Deep learning-based instance segmentation on 3D laser triangulation data for inline monitoring of particle size distributions in construction and demolition waste recycling

Resources Conservation and Recycling, vol. 205, pp. 107541

Abstract

Overlapping material flow presentations in construction and demolition waste (CDW) recycling make an inline particle size distribution (PSD) monitoring challenging. Here, we aim to build a deep-learning-based segmentation model for overlapping particles in 3D-laser-triangulation images of CDW. Our model was trained on three specially designed datasets with two transfer learning processes. U-net was employed as the backbone and Multi-Star algorithm was used to describe particle shapes. The final model demonstrated an impressive performance on test set, with a mean average precision (mAP) of 92.8% at IoU= 0.5. Comparing with the traditional segmentation algorithm based on image processing methods, the mAP can only reach to 27.4% on the same images. The shown model performance paves the way toward novel sensor technology applications for real-time PSD monitoring in CDW recycling.

Authors 3

1. X. Wu Aachen Department of Anthropogenic Material Cycles

   RWTH Aachen University

   Affiliation as printed

   Department of Anthropogenic Material Cycles, RWTH Aachen University, Wuellnerstr. 2, D-52062 Aachen, Germany

2. Nils Kroell corresponding Aachen Department of Anthropogenic Material Cycles

   RWTH Aachen University

   Affiliation as printed

   Department of Anthropogenic Material Cycles, RWTH Aachen University, Wuellnerstr. 2, D-52062 Aachen, Germany

3. Kathrin Greiff Aachen Department of Anthropogenic Material Cycles

   RWTH Aachen University

   Affiliation as printed

   Department of Anthropogenic Material Cycles, RWTH Aachen University, Wuellnerstr. 2, D-52062 Aachen, Germany