Ensemble Learning techniques for object detection in high-resolution satellite images
This work addresses object detection for defense applications using VHR imagery, but it is incremental as it applies existing ensembling methods to a new domain.
The study reviewed ensemble learning techniques for object detection in very high-resolution satellite images, demonstrating their value with a vehicle detection use-case in desert areas.
Ensembling is a method that aims to maximize the detection performance by fusing individual detectors. While rarely mentioned in deep-learning articles applied to remote sensing, ensembling methods have been widely used to achieve high scores in recent data science com-petitions, such as Kaggle. The few remote sensing articles mentioning ensembling mainly focus on mid resolution images and earth observation applications such as land use classification, but never on Very High Resolution (VHR) images for defense-related applications or object detection.This study aims at reviewing the most relevant ensembling techniques to be used for object detection on very high resolution imagery and shows an example of the value of such techniques on a relevant operational use-case (vehicle detection in desert areas).