A Semantic Segmentation Approach to River Sediment Identification

Ahmad Zaky Iswani; Kristiawan Dwi Usmanto

doi:10.25170/jurnalelektro.v19i2.8271

Authors

Ahmad Zaky Iswani Department of Mechanical Engineering, College of Engineering, Southern Taiwan University of Science and Technology, Tainan 71005, TAIWAN
Kristiawan Dwi Usmanto Department of Electrical Engineering, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember, Surabaya 60111, INDONESIA https://orcid.org/0009-0002-8879-5399

DOI:

https://doi.org/10.25170/jurnalelektro.v19i2.8271

Keywords:

Feature Pyramid Network, LinkNet, ResNet-50, River Sediment Identification, Semantic Segmentation, U-Net

Abstract

This study investigates the use of semantic segmentation to identify river sedimentation as part of soil erosion monitoring. Three neural network architectures, namely U-Net, LinkNet, and Feature Pyramid Network (FPN), were evaluated using 2,800 non-overlapping patches generated from 100 images of sediment formations from the IIUM River. After patch extraction, the image-mask pairs were divided into training and validation subsets using a 75:25 ratio, producing 2,100 training patches and 700 validation patches. Based on sediment-class Intersection over Union (IoU) under patch-level validation, U-Net achieved 0.8345, followed by LinkNet at 0.8189 and FPN at 0.2039. U-Net and LinkNet identified relatively uniform sediment formations more successfully, whereas all models struggled with irregular formations or sediment that was visually similar to river water. Because the data split was performed at the patch level, patches originating from the same source photograph may have been present in both the training and validation subsets. Therefore, the reported results should be interpreted as internal patch-level validation rather than as an estimate of generalisation to completely unseen source river images.

References

[1] X. Ren, C. Huang, Z. Zhu, L. Wang, N. Wang, X. Mu, D. Luo, J. Zhang, L. Xiang and X. Huang, "Anthropogenic forest degradation and its impact on soil erosion in the mountainous region of northern China," Catena, pp. 1-11, 2024.

[2] J. M. Molina, J. P. Llerena, L. Usero and M. A. Patricio, "Advances in instance segmentation: Technologies, metrics and applications in computer vision," Neurocomputing, vol. 625, pp. 1-19, 2025.

[3] M. Hasan, A. Khatun, S. A. Juie, S. B. Shorif and M. S. Uddin, "Deep Learning Based Semantic Image Segmentation: A Problem Driven Analysis of Architectures, Datasets, and Open Challenges," Journal of Computer and Communications, vol. 14, no. 7, pp. 12-45, 2026.

[4] S. Sharma, S. Sharma, and A. Athaiya, “Activation Functions in Neural Networks,” International Journal of Engineering Applied Sciences and Technology, vol. 4, no. 12, pp. 310–316, 2020, doi:10.33564/IJEAST.2020.v04i12.054

[5] H. Huang, G. Ho, V. Tang, E. Chow, W. Ip and K. Yung, "A lightweight fully convolutional neural network for efficient and accurate tunnel crack detection," Journal of Advanc ed Research, pp. 1-15, 2025.

[6] A. Oğuz and Ö. F. Ertuğrul, "Chapter 1 - Introduction to deep learning and diagnosis in medicine," in Diagnostic Biomedical Signal and Image Processing Applications with Deep Learning Methods, Intelligent Data-Centric Systems, 2023, pp. 1-40.

[7] A. Dastgir, W. Bin, M. U. Saeed, J. Sheng, L. Site and H. Hassan, "Attention LinkNet-152: a novel encoder-decoder based deep learning network for automated spine segmentation," Scientific Reports, pp. 1-13, 2025.

[8] Y. Chen, X. Zhu , Y. Li, Y. Wei and L. Ye, "Enhanced semantic feature pyramid network for small object detection," Signal Processing: ImageCommunication, vol. 113, pp. 1-10, 2023.

[9] J. Wan, Z. Qin, X. Cui, F. Yang, M. Yasir, B. Ma and X. Liu, "MBES Seabed Sediment Classification Based on a Decision Fusion Method Using Deep Learning Model," Remote Sensing, vol. 14, no. 3708, pp. 1-22, 2022.

[10] R. V. Poursorkhabi, "Data-driven study of flow & sediment in circular channels using GPR & ANFIS," Multiscale and Multidisciplinary Modeling, Experiments and Design, pp. 1-22, 2026.

[11] Q. Zhu, Y. Zheng, Y. Jiang and J. Yang, "Efficient Multi-Class Semantic Segmentation of High Resolution Aerial Imagery With Dilated Linknet," in IEEE International Symposium on Geoscience and Remote Sensing (IGARSS), 2019.

[12] M. Wu, Z. Shu, J. Zhang and X. Hu, "HRLINKNET: Linknet With High-Resolution Representation for High-Resolution Satellite Imagery," in 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2021.

[13] M. P. M. M, V. Mehrotra, S. Aiyar, U. Verma and R. M. Pai, "Automatic Segmentation of River and Land in SAR Images: A Deep Learning Approach," in 2019 IEEE Second International Conference on Artificial Intelligence and Knowledge Engineering (AIKE), 2019.

[14] N. A. Muhadi, A. F. Abdullah, S. K. Bejo, M. R. Mahadi and A. Mijic, "Deep Learning Semantic Segmentation for Water Level Estimation Using Surveillance Camera," Applied Sciences, vol. 11, no. 9691, pp. 1-25, 2021.

[15] M. N. Mahmud, M. K. Osman, A. P. Ismail, F. Ahmad, K. A. Ahmad and A. Ibrahim, "Road Image Segmentation using Unmanned Aerial Vehicle Images and DeepLab V3+ Semantic Segmentation Model," in 2021 11th IEEE International Conference on Control System, Computing and Engineering (ICCSCE 2021), Penang, Malaysia, 2021.

[16] A. Tsellou, K. Moirogiorgou, G. Plokamakis, G. Livanos, K. Kalaitzakis and M. Zervakis, "Aerial video inspection of Greek power lines structures using machine learning techniques," in 2022 IEEE International Conference on Imaging Systems and Techniques (IST), Kaohsiung, Taiwan, 2022.

[17] N. Rania, H. Douzi, L. Yves and T. Sylvie, "Semantic Segmentation of Diabetic Foot Ulcer Images: Dealing with Small Dataset in DL Approaches," Image and Signal Processing, pp. 162-169, 2020.