UPM Institutional Repository

Res-UNet ensemble learning for mineral optical microscopy images semantic segmentation


Citation

Abdul Halin, Alfian and Jiang, Chong and Perumal, Thinagaran and Manshor, Noridayu and Abdullah, Lili Nurliyana and Yang, Baohua (2024) Res-UNet ensemble learning for mineral optical microscopy images semantic segmentation. Minerals, 14 (12). art. no. 1281. pp. 1-20. ISSN 2075-163X

Abstract

Abstract: In geology and mineralogy, optical microscopic images have become a primary research focus for intelligent mineral recognition due to their low equipment cost, ease of use, and distinct mineral characteristics in imaging. However, due to their close reflectivity or transparency, some minerals are not easily distinguished from other minerals or background. Secondly, the number of background pixels often vastly exceeds the number of pixels for individual mineral particles, and the number of pixels of different mineral particles in the image also varies significantly. These have led to the issue of data imbalance. This imbalance results in lower recognition accuracy for categories with fewer samples. To address these issues, a flexible ensemble learning for semantic segmentation based on multiple optimized Res-UNet models is proposed, introducing dice loss and focal loss functions and incorporating a pre-positioned spatial transformer networks block. Twelve optimized Res-UNet models were used to construct multiple Res-UNet ensemble learnings using heterogeneous ensemble strategies. The results demonstrate that the system integrated with five learners using the weighted voting fusion method (RUEL-5-WV) achieved the best performance with a mean Intersection over Union (mIOU) of 91.65 across all nine categories and an IOU of 84.33 for the transparent mineral (gangue). The results indicate that this ensemble learning scheme outperforms individual optimized Res-UNet models. Compared to the classical Deeplabv3 and PSPNet, this scheme also exhibits significant advantages.


Download File

[img] Text
118276.pdf - Published Version
Available under License Creative Commons Attribution.

Download (10MB)
Official URL or Download Paper: https://www.mdpi.com/2075-163X/14/12/1281

Additional Metadata

Item Type: Article
Divisions: Faculty of Computer Science and Information Technology
DOI Number: https://doi.org/10.3390/min14121281
Publisher: Multidisciplinary Digital Publishing Institute
Keywords: Optical microscopy images; Deep learning; Ensemble learning; Semantic segmentation; Mineralogy
Depositing User: Ms. Che Wa Zakaria
Date Deposited: 03 Jul 2025 02:57
Last Modified: 03 Jul 2025 02:57
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.3390/min14121281
URI: http://psasir.upm.edu.my/id/eprint/118276
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item