Mask R-CNN Applied to Quasi-particle Segmentation from the Hybrid Pelletized Sinter (HPS) Process

Natália F. De C. Meira, Mateus C. Silva, Andrea G. C. Bianchi, Cláudio B. Vieira, Alinne Souza, Efrem Ribeiro, Roberto O. Junior, Ricardo A. R. Oliveira

2022

Abstract

Particle size is an important quality parameter for raw materials in steel industry processes. In this work, we propose to implement the Mask-R-CNN algorithm to segment quasi-particles by size classes. We created a dataset with real images of an industrial environment, labeled the quasi-particles by size classes, and performed four training sessions by adjusting the model’s hyperparameters. The results indicated that the model segments with well-defined edges and applications as classes correctly. We obtained a mAP between 0.2333 and 0.2585. Additionally, hit and detection rates increase for larger particle size classes.

Paper Citation

in Harvard Style

Meira N., Silva M., Bianchi A., Vieira C., Souza A., Ribeiro E., O. Junior R. and Oliveira R. (2022). Mask R-CNN Applied to Quasi-particle Segmentation from the Hybrid Pelletized Sinter (HPS) Process. In Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP; ISBN 978-989-758-555-5, SciTePress, pages 462-469. DOI: 10.5220/0010836900003124

in Bibtex Style

@conference{visapp22,
author={Natália F. De C. Meira and Mateus C. Silva and Andrea G. C. Bianchi and Cláudio B. Vieira and Alinne Souza and Efrem Ribeiro and Roberto O. Junior and Ricardo A. R. Oliveira},
title={Mask R-CNN Applied to Quasi-particle Segmentation from the Hybrid Pelletized Sinter (HPS) Process},
booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP},
year={2022},
pages={462-469},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010836900003124},
isbn={978-989-758-555-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 4: VISAPP
TI - Mask R-CNN Applied to Quasi-particle Segmentation from the Hybrid Pelletized Sinter (HPS) Process
SN - 978-989-758-555-5
AU - Meira N.
AU - Silva M.
AU - Bianchi A.
AU - Vieira C.
AU - Souza A.
AU - Ribeiro E.
AU - O. Junior R.
AU - Oliveira R.
PY - 2022
SP - 462
EP - 469
DO - 10.5220/0010836900003124
PB - SciTePress