Implementation of U-Net for Aerial Imagery Segmentation

The development of environmental mapping and monitoring technology has been advancing rapidly in recent years, thanks to improvements in remote sensing and drone technology. Drones have become a valuable tool for collecting high-resolution aerial imagery data quickly and efficiently. This data has numerous applications, including image segmentation, environmental monitoring, and natural resource management. One of the main challenges in using aerial imagery is object segmentation, which involves separating specific objects from the background of the image. This research focuses on implementing and evaluating the U-Net model for object segmentation of buildings, rice fields, and runways in aerial imagery. Using aerial imagery data from Google Earth, the study explores various technical aspects of the U-Net model, such as data preprocessing, model training, and performance evaluation. The results demonstrate that the U-Net model is capable of accurately and efficiently segmenting buildings, rice fields, and runways. This research has the potential to significantly contribute to the field of aerial imagery segmentation and expand the practical applications of the U-Net model in aerial resource monitoring and management scenarios.