From Pixels to Planes: Minimum Ground Sample Distance for Aircraft
Matthew Ciolino, Willie Maddox
TL;DR
The paper investigates the minimum Ground Sample Distance (GSD) necessary to reliably detect aircraft in remote-sensing imagery, using the AllPlanes 120 dataset and a YOLOv8s detector trained on the finest-resolution data. By downsampling 640×640 images to six GSD levels ($2.4$–$0.38$ m) and evaluating with $mAP_{50-95}$, it shows that a GSD of $0.86$ m is typically sufficient for most aircraft, with smaller targets (<$20$ m wingspan) requiring finer resolution. The study further analyzes performance by wingspan, identifies dominant error types via TIDE, and discusses practical implications for lightweight, high-altitude optical systems. The findings inform camera design trade-offs between weight and detection accuracy and highlight potential enhancements from sensor tech and image restoration methods. This work provides actionable guidance for deploying aircraft detection systems on platforms where payload and power constraints limit achievable GSDs.
Abstract
This study investigates the impact of ground sample distance (GSD) on the detection performance of various sized aircraft using the proprietary AllPlanes 120 dataset. The data set comprises 120 civilian, military and museum aircraft from multiple satellite/aerial sources collected over two years. Resolutions ranging from 2.4 to 0.3 meters GSD were simulated. Performance metrics were derived from a YOLOv8s model trained on down-sampled versions of zoom level 19 (0.3m GSD) imagery. The results indicate that a GSD of at least 0.86m is required to accurately detect most aircraft, particularly those with wingspans shorter than 20 meters. Due to weight constraints in high-altitude platforms, this GSD specification can inform camera design to minimize weight while maintaining detection accuracy.