ORIGINAL PAPER
RA-UNet++ based image segmentation of adherent ores
1
School of Mechanical and Electrical Engineering, jiangxi University of Science and Technology Ganzhou, jiangxi Province, China
2
Jiangxi Provincial Key Laboratory of Particle Technology, Nanchang , Jiangxi 330013, China
3
School of Mechanical and Electrical Engineering, jiangxi University of Science and TechnologyGanzhou, jiangxi Province, China
Submission date: 2025-06-16
Final revision date: 2025-07-03
Acceptance date: 2025-09-05
Publication date: 2026-04-15
Corresponding author
Xiaoyan Luo
School of Mechanical and Electrical Engineering, jiangxi University of Science and Technology Ganzhou, jiangxi Province, China
School of Mechanical and Electrical Engineering, jiangxi University of Science and Technology Ganzhou, jiangxi Province, China
Gospodarka Surowcami Mineralnymi – Mineral Resources Management 2026;42(2):69-87
KEYWORDS
TOPICS
ABSTRACT
Ore particle size information is an important indicator for evaluating the operating status and production efficiency of crushers. However, in the actual industrial environment, adhesion phenomena often occur during the acquisition and transportation of ores, resulting in overlapping edges and blurred contours of ores in the images. Traditional image segmentation methods are difficult to achieve high-precision recognition and segmentation. To this end, this paper proposes an RA-UNet++ adhered ore image segmentation method based on the improved UNet++ structure to improve the segmentation performance in complex scenes. Based on the UNet++ coender-decoding architecture, the residual module and the self-attention mechanism are integrated to enhance the model’s ability to extract and express the edge details of ores. Meanwhile, multi-scale atrous convolution is introduced at the end of the encoder to construct the Atrous Spatial Pyramid Pooling (ASPP) structure, expand the receptive field, enhance the multi-scale perception ability for ores of different particle sizes, and thereby improve the overall segmentation effect. The experimental results show that RA-UNet++ performs excellently in the task of image segmentation of adhered ores, significantly improving the clarity and integrity of edge segmentation. Compared with the Otsu method and the standard UNet model, this method has advantages in terms of robustness, boundary preservation, and segmentation accuracy. The pixel-level and target-level segmentation accuracy rates of adhered ore images both exceed 93%, showing good potential for industrial applications.
CONFLICT OF INTEREST
The Authors have no conflict of interest to declare.
METADATA IN OTHER LANGUAGES:
Polish
Segmentacja obrazów rud przylegających z wykorzystaniem sieci RA-UNet++
segmentacja obrazu, UNet++, moduł resztkowy, mechanizm samo-uwagi, Atrous Spatial Pyramid Pooling
Informacje dotyczące wielkości cząstek rudy stanowią ważny wskaźnik służący do oceny stanu pracy i wydajności produkcyjnej kruszarek. Jednak w rzeczywistych warunkach przemysłowych podczas pozyskiwania i transportu rud często dochodzi do zjawisk zlepiania się cząstek, co powoduje nakładanie się krawędzi i rozmycie konturów rudy na obrazach. Tradycyjne metody segmentacji obrazów mają trudności z osiągnięciem wysokiej precyzji rozpoznawania i segmentacji. W związku z tym w niniejszym artykule proponuje się metodę segmentacji obrazów rudy ze zjawiskiem przywierania RA-UNet++, opartą na ulepszonej strukturze UNet++, mającą na celu poprawę wydajności segmentacji w złożonych scenach. W oparciu o architekturę kodowania i dekodowania UNet++ zintegrowano moduł resztkowy oraz mechanizm samo-uwagi, aby wzmocnić zdolność modelu do wyodrębniania i wyrażania szczegółów krawędzi rudy. Jednocześnie na końcu kodera wprowadzono wieloskalową konwolucję atrous, aby zbudować strukturę Atrous Spatial Pyramid Pooling (ASPP), rozszerzyć pole receptywne, zwiększyć zdolność postrzegania w wielu skalach dla rud o różnych rozmiarach cząstek, a tym samym poprawić ogólny efekt segmentacji. Wyniki eksperymentalne pokazują, że RA-UNet++ doskonale radzi sobie z zadaniem segmentacji obrazów rud przylegających, znacznie poprawiając klarowność i integralność segmentacji krawędzi. W porównaniu z metodą Otsu i standardowym modelem UNet metoda ta ma przewagę pod względem odporności, zachowania granic i dokładności segmentacji. Wskaźniki dokładności segmentacji obrazów rudy przylegającej, zarówno na poziomie pikseli, jak i na poziomie obiektów docelowych, przekraczają 93%, co wskazuje na duży potencjał w zastosowaniach przemysłowych.
REFERENCES (29)
1.
Badrinarayanan et al. 2017 – Badrinarayanan, V., Kendall, A. and Cipolla, R. 2017. Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE transactions on pattern analysis and machine intelligence 39(12), https://doi.org/10.1109/TPAMI.....
2.
Chaurasia et al. 2025 – Chaurasia, A., Sharma, V. S., Srivastava, A., Abbas, Z. and Nagpal, R. 2025. Insights into Brain Tumor Detection: Exploring U-Net Architecture. [In:] Mathematics and Logics in Computer Science, pp. 173–183, https://doi.org/10.1007/978-98....
3.
Chen, H. 2025. Applications of deep learning in medical image segmentation: an evaluation of U-Net architecture and its enhancements. International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2024), pp. 728–735, https://doi.org/10.1117/12.306....
4.
Dai et al. 2017 – Dai, J., Qi, H., Xiong, Y., Li, Y., Zhang, G., Hu, H. and Wei, Y. 2017. Deformable convolutional networks. Proceedings of the IEEE international conference on computer vision, pp. 764–773, https://doi.org/10.48550/arXiv....
5.
Dash et al. 2017 – Dash, M., Londhe, N.D., Ghosh, S., Semwal, A. and Sonawane, R.S. 2019. PsLSNet: Automated psoriasis skin lesion segmentation using modified U-Net-based fully convolutional network. Biomedical Signal Processing and Control 52(12), pp. 226–237, https://doi.org/10.1016/j.bspc....
6.
He et al. 2016 – He, K., Zhang, X., Ren, S. and Sun, J. 2016. Identity mappings in deep residual networks. Lecture Notes in Computer Science 9908, pp. 630–645). https://doi.org/10.1007/978-3-....
7.
Huang et al. 2022 – Huang, M., Liu, Y. and Yang, Y. 2022. Edge detection of ore and rock on the surface of explosion pile based on improved Canny operator. Alexandria Engineering Journal 61(12), pp. 10769–10777, https://doi.org/10.1016/j.aej.....
8.
Jin et al. 2021 – Jin, F., Zhan, K., Chen, S., Huang, S. and Zhang, Y. 2021. Image segmentation method of mine pass soil and ore based on the fusion of the confidence edge detection algorithm and mean shift algorithm. Gospodarka Surowcami Mineralnymi – Mineral Resources Management 37(4), pp. 133–152, https://doi.org/10.24425/gsm.2....
9.
Krizhevsky et al. 2012 – Krizhevsky, A., Sutskever, I. and Hinton, G.E. 2012. ImageNet classification with deep convolutional neural networks. Advances in neural information processing systems 25.
10.
Li et al. 2023 – Li, F., Liu, X., Yin, Y. and Li, Z. 2023. DDR-Unet: A high-accuracy and efficient ore image segmentation method. IEEE Transactions on Instrumentation and Measurement 99, https://doi.org/10.1109/TIM.20....
11.
Long et al. 2015 – Long, J., Shelhamer, E. and Darrell, T. 2015. Fully convolutional networks for semantic segmentation. Conference: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3431–3440, https://doi.org/10.1109/CVPR.2....
12.
Mac et al. 2019 – Mac, K.N.C., Joshi, D., Yeh, R.A., Xiong, J., Feris, R.S. and Do, M.N. 2019. Learning motion in feature space: Locally-consistent deformable convolution networks for fine-grained action detection. IEEE/CVF International Conference on Computer Vision, pp. 6282–6291.
13.
Mustafa et al. 2020 – Mustafa, N., Zhao, J., Liu, Z., Zhang, Z. and Yu, W. 2020. Iron ORE region segmentation using high-resolution remote sensing images based on Res-U-Net. IGARSS 2020–2020 IEEE International Geoscience and Remote Sensing Symposium, pp. 2563–2566, https://doi.org/10.1109/IGARSS....
14.
Naylor et al. 2018 – Naylor, P., Laé, M., Reyal, F. and Walter, T. 2018. Segmentation of nuclei in histopathology images by deep regression of the distance map. IEEE transactions on medical imaging 38(2), pp. 448–459, https://doi.org/10.1109/TMI.20....
15.
Oktay et al. 2018 – Oktay, O., Schlemper, J., Folgoc, L.L., Lee, M., Heinrich, M.P., Misawa, K., Mori, K., McDonagh, S., Hammerla, N.Y., Kainz, B., Glocker, B. and Rueckert, D. 2018. Attention u-net: Learning where to look for the pancreas, https://doi.org/10.48550/arXiv....
16.
Ronneberger et al. 2015 – Ronneberger, O., Fischer, P. and Brox, T. 2015. U-net: Convolutional networks for biomedical image segmentation. Conference: International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 234–241, https://doi.org/10.1007/978-3-....
17.
Salinas et al. 2005 – Salinas, R.A., Raff, U. and Farfan, C. 2005. Automated estimation of rock fragment distributions using computer vision and its application in mining. IEE Proceedings-Vision, Image and Signal Processing 152(1), https://doi.org/10.1049/ip-vis....
18.
Shaw et al. 2018 – Shaw, P., Uszkoreit, J. and Vaswani, A. 2018. Self-attention with relative position representations, https://doi.org/10.48550/arXiv....
19.
Shrivastava et al. 2023 – Shrivastava, S., Bhattacharjee, S. and Deb, D. 2023. Segmentation of mine overburden dump particles from images using Mask R CNN. Scientific Reports 13(1), https://doi.org/10.1038/s41598....
20.
Song et al. 2019 – Song, W., Zheng, N., Liu, X., Qiu, L. and Zheng, R. 2019. An improved u-net convolutional networks for seabed mineral image segmentation. Ieee Access 7, pp. 82744–82752, https://doi.org/10.1109/ACCESS....
21.
Sun et al. 2022 – Sun, G., Huang, D., Cheng, L., Jia, J., Xiong, C. and Zhang, Y. 2022. Efficient and lightweight framework for real-time ore image segmentation based on deep learning. Minerals 12(5), https://doi.org/10.3390/min120....
22.
Tavakolian et al. 2025 – Tavakolian, N., Nazemi, A. and Suen, C.Y. 2025. Review of the AI-Based Analysis of Abdominal Organs from Routine CT Scans. Applied Sciences 15(5), https://doi.org/10.3390/app150....
23.
Touvron et al. 2021 – Touvron, H., Cord, M., Sablayrolles, A., Synnaeve, G. and Jégou, H. 2021. Going deeper with image transformers. Conference: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), pp. 32–42, https://doi.org/10.1109/ICCV48....
24.
Wang et al. 2023 – Wang, W., Li, Q., Zhang, D. and Fu, J. 2023. Image segmentation of adhesive ores based on MSBA-Unet and convex-hull defect detection. Engineering Applications of Artificial Intelligence 123, https://doi.org/10.1016/j.enga....
25.
Xiao et al. 2020 – Xiao, D., Liu, X., Le, B. T., Ji, Z. and Sun, X. 2020. An ore image segmentation method based on RDU-Net model. Sensors 20(17), https://doi.org/10.3390/s20174....
26.
Yuan, L. and Duan, Y. 2018. A method of ore image segmentation based on deep learning. [In:] Intelligent Computing Methodologies, pp. 508–519, https://doi.org/10.1007/978-3-....
27.
Yurtkulu et al. 2019 – Yurtkulu, S.C., Şahin, Y.H. and Unal, G. 2019. Semantic segmentation with extended DeepLabv3 architecture. 2019 27th signal processing and communications applications conference (SIU), https://doi.org/10.1109/SIU.20....
28.
Zhang et al. 2017 – Zhang, G., Li, M., Zhan, Y. and Shi, X. 2017. Ore image thresholding segmentation using double windows with fisher discrimination. 2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), pp. 2715-2719, https://doi.org/10.1109/FSKD.2....
29.
Zhou et al. 2019 – Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N. and Liang, J. 2019. Unet++: Redesigning skip connections to exploit multiscale features in image segmentation. IEEE transactions on medical imaging 39(6), pp. 1856–1867, https://doi.org/10.48550/arXiv....
| eISSN: | 2299-2324 |
| ISSN: | 0860-0953 |
© 2006-2026 Journal hosting platform by Bentus
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
