Hierarchical Multi-Modal Enhancement for Robust Transmission Line Detection

Ensuring a stable power supply in rural areas relies heavily on effective inspection of power equipment, particularly transmission lines (TLs). However, detecting TLs from aerial imagery can be challenging when dealing with misalignments between visible light (RGB) and infrared (IR) images, as well as mismatched high- and low-level features in convolutional networks. To address these limitations, we propose a Hierarchical Multi-Modal Enhancement Network (HMMEN) that integrates RGB and IR data for robust and accurate TL detection. Our method introduces two key components: 1) a Mutual Multi-Modal Enhanced Block (MMEB), which fuses and enhances hierarchical RGB and IR feature maps in a coarse-to-fine manner, and 2) a Feature Alignment Block (FAB) that corrects misalignments between decoder outputs and IR feature maps by leveraging deformable convolutions. We employ MobileNet-based encoders for both RGB and IR inputs to accommodate edge-computing constraints and reduce computational overhead. Experimental results on diverse weather and lighting conditions-fog, night, snow, and daytime-demonstrate the superiority and robustness of our approach compared to state-of-the-art methods, resulting in fewer false positives, enhanced boundary delineation, and better overall detection performance. This framework thus shows promise for practical large-scale power line inspections with uncrewed aerial vehicles. Note to Practitioners - This paper targets to address transmission line detection by introducing a lightweight hybrid-modality mutual-enhancement network (HMMEN) that balances accuracy and inference speed for UAV deployment. Unlike conventional fusion strategies that neglect inter-modal misalignment, HMMEN explicitly models cross-modal complementarity and rectifies cross-scale discrepancies between high- and low-level features. Our experiments demonstrate competitive performance with minimal latency, paving the way for future onboard UAV applications.