Multi-Modal Semantic Segmentation of Electrolyzer Components for Sustainable Hydrogen Technologies: A Dual-Branch Deep Learning Approach
Not on the map. Why?
Abstract (arXiv)
Accurate segmentation of electrolyzer materials is essential for automated disassembly, sustainable recycling, and circular manufacturing in hydrogen technologies. However, this task is challenging due to strong visual similarity between materials, spectral overlap, irregular shapes, and severe class imbalance. To address these challenges, we propose an AI-driven dual-branch framework, Hyperspectral-RGB Electrolyzer Materials Network (HREM-Net), that combines hyperspectral imaging (HSI) and RGB images for electrolyzer material segmentation. We implemented several innovative modules, including Efficient Channel Attention, Coordinate Attention, Mobile Inverted Bottleneck blocks, and Atrous Spatial Pyramid Pooling to capture spectral and spatial features from HSI, and RGB images. With an adaptive gated cross-modal fusion module and composite loss function, HREM-Net achieves a mean class accuracy of 91.66% and a mean Intersection over Union (mIoU) of 0.82 on the Electrolyzers-HSI dataset, outperforming baseline segmentation models. Cross-dataset validation on the PCB-Vision dataset demonstrates strong generalization with 96.91% accuracy and 0.93 mIoU. This work poses its potential as an industrial application to improve electrolyzer efficiency, thereby improving the predictive maintenance of hydrogen production.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 24 of its 53 references that are in the map. First posted on arXiv 17 Jul 2026.
Builds on
24 of 53 references.
References · 24
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
- 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,223
- 2021TransUNet: Transformers Make Strong Encoders for Medical Image Segmentationcited by 1,558
- 2018UNet++: A Nested U-Net Architecture for Medical Image Segmentationcited by 1,525
- 2018Dual Attention Network for Scene Segmentationcited by 1,003
- 2019ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networkscited by 575
- 2014Deeply-Supervised Netscited by 367
- 2021Coordinate Attention for Efficient Mobile Network Designcited by 241
- 2017Tversky loss function for image segmentation using 3D fully convolutional deep networkscited by 228
- 2021Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentationcited by 88
- 2018RedNet: Residual Encoder-Decoder Network for indoor RGB-D Semantic Segmentationcited by 66
12 more
- 2022PolyLoss: A Polynomial Expansion Perspective of Classification Loss Functionscited by 63
- 2023Cross-Modal Retrieval: A Systematic Review of Methods and Future Directionscited by 40
- 2021Expansion-Squeeze-Excitation Fusion Network for Elderly Activity Recognitioncited by 23
- 2021Hyperspectral Image Denoising and Anomaly Detection Based on Low-rank and Sparse Representationscited by 7
- 2024Generative AI in Industrial Machine Vision -- A Reviewcited by 5
- 2021Hyperspectral Band Selection for Multispectral Image Classification with Convolutional Networkscited by 4
- 2024PCB-Vision: A Multiscene RGB-Hyperspectral Benchmark Dataset of Printed Circuit Boardscited by 3
- 2024Hyperspectral Dataset and Deep Learning methods for Waste from Electric and Electronic Equipment Identification (WEEE)cited by 3
- 2021Domestic waste detection and grasping points for robotic picking upcited by 1
- 2024WasteGAN: Data Augmentation for Robotic Waste Sorting through Generative Adversarial Networkscited by 1
- 2025Electrolyzers-HSI: Close-Range Multi-Scene Hyperspectral Imaging Benchmark Datasetcited by 1
- 2026An Ensemble Learning Approach towards Waste Segmentation in Cluttered Environmentcited by 1
Built on by
Not cited yet.