Data from the article "DTL-IceNet: A Dual-Task Learning Architecture with Multi-Scale Fusion Mechanisms for Enhanced Ice Detection on Transmission Lines"

Description of the data and file structure example_icing_images: This dataset contains images of transmission line icing of various types, used for ice layer recognition and thickness detection tasks. Ice_code: The above code is a semantic segmentation model based on the ground truth image of the power grid ice-covered line. By combining the deep learning model with the multimodal branch, the refined segmentation of the ice-covered contour is achieved. The nes folder contains the entire model structure and detailed codes of each module. For other usage methods, please refer to the README.