Classification of Aluminum Scrap by Laser Induced Breakdown Spectroscopy (LIBS) and RGB D Image Fusion Using Deep Learning Approaches

22 Pages Posted: 9 Nov 2022 Last revised: 1 Feb 2023

Abstract

This study presents two novel methods for fusing RGB and Depth images with LIBS using Deep Learning models. The first method is a single-output model that combines LIBS UNET and two DenseNets in a late fusion framework. The second method is a multiple-output model that uses the structure of the single-output model to enhance learning and avoid overfitting. The first sorting task is separating Cast and Wrought (C&W) aluminum. The second is the division of the post-consumer aluminum scrap into three commercially interesting fractions. The single-output model performs best for separating C&W, with a Precision, Recall, and F1-score of 99%. The multiple-output model performs best for classifying the three selected commercial fractions, with a Precision, Recall, and F-score of 86%, 83%, and 84%. The presented data fusion method for LIBS and computer vision images encompasses great potential for sorting post-consumer aluminum scrap.

Keywords: artificial intelligence, metal recycling, Data Fusion, Deep Learning Computer Vision, Laser-Induced Breakdown Spectroscopy (LIBS), Multiple-loss fusion

Suggested Citation

Diaz-Romero, Dillam Jossue and Van den Eynde, Simon and Zaplana, Isiah and Zhou, Chuangchuang and Sterkens, Wouter and Goedemé, Toon and Peeters, Jef, Classification of Aluminum Scrap by Laser Induced Breakdown Spectroscopy (LIBS) and RGB D Image Fusion Using Deep Learning Approaches. Available at SSRN: https://ssrn.com/abstract=4272447 or http://dx.doi.org/10.2139/ssrn.4272447

Simon Van den Eynde

KU Leuven ( email )

Oude Markt 13
Leuven, 3000
Belgium

Isiah Zaplana

KU Leuven ( email )

Chuangchuang Zhou

KU Leuven ( email )

Wouter Sterkens

KU Leuven ( email )

Toon Goedemé

KU Leuven ( email )

Oude Markt 13
Leuven, 3000
Belgium

Jef Peeters

KU Leuven ( email )

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
107
Abstract Views
624
Rank
448,824
PlumX Metrics