Transmission Line Icing Thickness Prediction Dataset

In the energy sector, transmission line icing is a crucial factor affecting power supply. The current challenge is the lack of detailed monitoring data, making it difficult to take timely measures to prevent line damage. Existing solutions mostly rely on weather data for predictions, but their accuracy is insufficient. This dataset aims to more accurately predict icing thickness through image data, meeting the timely decision-making needs of the power sector. Data collection is carried out by high-resolution cameras under extreme weather conditions, ensuring coverage of different icing levels. For quality control, multiple rounds of annotation and consistency checks are conducted and reviewed by experts in the power field. The annotation team includes 20 experienced technicians. Data preprocessing includes image enhancement, denoising, and resolution adjustment, stored in JPG format and organized by time and geographic location.