# Feasibility Study of Scrap Grading Systems Based on Three-Dimensional Vision Technology

## TL;DR

This paper explores using 3D vision technology to improve the fairness and efficiency of scrap grading by achieving high accuracy in object reconstruction and segmentation.

## Contribution

The study introduces a 3D vision-based workflow for scrap grading with a novel comparison of point cloud segmentation methods.

## Key findings

- The multi-view 3D reconstruction algorithm achieves 1 mm accuracy in scrap grading.
- Euclidean clustering outperforms other methods with a 99.35% mIoU score and 0.5 mm thickness error.
- The 3D workflow shows better robustness than 2D image-based methods for thickness inference.

## Abstract

To address the inefficiency and unfairness of traditional manual scrap sorting, we propose the application of 3D vision technology for grading in this work. The multi-view 3D reconstruction algorithm achieves an accuracy within 1 mm in both synthetic and real scrap scenes. This level of accuracy meets the requirements for scrap grading. Subsequently, an automated processing workflow in a non-overlapping scrap scenario is investigated, in which a pipeline based on the multi-view reconstruction integrating point cloud segmentation technique is proposed. Four-point cloud clustering segmentation methods, including Euclidean clustering, Kmeans, DBSCAN and Region Grow, are compared, and it is found that the Euclidean-clustering-based point cloud segmentation algorithm provides the best overall trade-off, achieving an mIoU score of 99.35%, while the thickness measurement error is less than 0.5 mm. The workflow suggests improved robustness and reliability compared to using a single 2D image for thickness inference. These results indicate that 3D vision may provide a valuable basis for the future development of scrap grading systems.