STC-Net: Electroluminescence-Based Solar Cell Crack Segmentation for Power Loss Estimation

2026-08-03

AI Summary

This study addresses the challenge of accurately segmenting thin, structurally constrained cracks in electroluminescence (EL) images of solar cells—a critical bottleneck in photovoltaic reliability analysis and power loss estimation. To this end, we propose STC-Net, the first framework to jointly model crack segmentation and power loss prediction. By integrating edge priors, spectral priors, and a boundary-topology refinement module, STC-Net enhances boundary precision while preserving structural continuity of cracks. The segmentation output further informs a proxy for inactive cell areas, enabling direct estimation of power loss. Evaluated on the PVEL-S dataset, our model achieves 95.98 MIoU and 98.01 MDice during training, and demonstrates strong generalization with 72.52 MIoU and 80.16 MDice on the test set, validating its effectiveness and robustness.

Abstract

Accurate crack assessment in electroluminescence (EL) images is important for photovoltaic (PV) reliability analysis, yet existing segmentation methods often fail to capture the thin, elongated, and structurally constrained nature of crack defects. This paper proposes a Solar Topology Crack Network (STC-Net) that incorporates edge priors, spectral priors, and a boundary-topology refinement module to improve crack continuity and boundary preservation. The framework further extends segmentation to power-loss estimation by deriving a crack-associated inactive-area proxy from the predicted masks. Experiments on the PVEL-S dataset show that STC-Net achieves 95.98 MIoU, 98.01 MDice, and 98.00 MAcc during training, and 72.52 MIoU and 80.16 MDice on unseen test samples. These results demonstrate that STC-Net provides accurate crack localization while offering a practical link between EL-based defect segmentation and PV degradation assessment.

Problem

• crack segmentation
• electroluminescence
• solar cell
• power loss estimation
• photovoltaic reliability

Innovation

• crack segmentation
• electroluminescence imaging
• topology-aware refinement
• power loss estimation
• solar cell defect analysis