Development of an AI-Based Detection of Nesting Boxes in the Overhead Line Network of a Transmission System Operator: An Application for Environmental and Operational Monitoring

This paper presents an innovative application of artificial intelligence (AI) for the automated detection of nest boxes within the overhead line network of a transmission system operator (TSO). An object detection model based on the You Only Look Once (YOLO) architecture was developed, leveraging its inherent efficiency and real-time processing capabilities. The methodology encompassed a rigorous data pipeline, including the acquisition and annotation of high-resolution image data and the implementation of data augmentation strategies to enhance model robustness and generalization performance. An empirical evaluation conducted on the validation dataset resulted in an F1-score of 99.33%, reflecting a very high detection accuracy. This research establishes a foundation for future investigations into related tasks, such as the automated detection of avian nests and the classification of bird species inhabiting these artificial nesting structures.