Performance Evaluation of YOLOv8 and YOLOv9 for Object Detection in Remote Sensing Images

Object detection in remote sensing images is a crucial technology with applications in various fields such as unmanned aerial vehicles, intelligent traffic monitoring, and aerospace. Deep learning techniques have shown promising results in the process of detecting objects, but challenges unique to remote sensing images, for instance, scale variation and complex backgrounds, remain. The purpose of this study is to analyze the differences between two state-of-the-art frameworks, the first being first YOLOv8, and the second is YOLOv9, regarding the domain of object detection. The evaluation focuses on two key performance metrics: processing time and training accuracy. The results reveal the superiority of YOLOv9 over YOLOv8 regarding the aspect of accuracy, demonstrating superior detection capabilities. Conversely, YOLOv8 exhibits notable advantages in terms of training and processing time, showcasing its efficiency and reduced computational burden. Upon the analysis of the findings, the selection of the most suitable model depends on the specific scenario and problem requirements. The outcomes of this study provide valuable insights for researchers and practitioners seeking to optimize object detection systems in remote sensing applications.