Training Set Effect on Super Resolution for Automated Target Recognition

TL;DR

This paper investigates how different training sets influence the effectiveness of Super Resolution Generative Adversarial Networks (SRGAN) in enhancing satellite imagery for automated target recognition, revealing that curated and diverse datasets improve performance.

Contribution

It systematically evaluates the impact of various land-use training sets on SRGAN performance for super resolution, classification, and detection tasks.

Findings

Curated training sets improve CV task performance.

Complex image distributions benefit object detection.

SR may have limited benefits for near-solved datasets.

Abstract

Single Image Super Resolution (SISR) is the process of mapping a low-resolution image to a high resolution image. This inherently has applications in remote sensing as a way to increase the spatial resolution in satellite imagery. This suggests a possible improvement to automated target recognition in image classification and object detection. We explore the effect that different training sets have on SISR with the network, Super Resolution Generative Adversarial Network (SRGAN). We train 5 SRGANs on different land-use classes (e.g. agriculture, cities, ports) and test them on the same unseen dataset. We attempt to find the qualitative and quantitative differences in SISR, binary classification, and object detection performance. We find that curated training sets that contain objects in the test ontology perform better on both computer vision tasks while having a complex distribution of images allows object detection models to perform better. However, Super Resolution (SR) might not be beneficial to certain problems and will see a diminishing amount of returns for datasets that are closer to being solved.