A real-time UAS hyperspectral anomaly detection system

TL;DR

The paper addresses the challenge of delivering timely, actionable anomaly information from hyperspectral imagery captured by a UAS. It introduces an end-to-end pipeline that performs real-time anomaly detection onboard using a global RX detector, transmits concise results via monitor-mode WiFi, and supports interactive ground-based analysis with real-time georectification through Spacecube. The system is demonstrated on Memphis and Uvalde test flights, achieving average cube processing times that meet real-time constraints and ROC-based detector performance comparable to prior work, while highlighting practical limitations in transmission, calibration, and processing. This work demonstrates a practical, low-cost path to immediate hyperspectral anomaly insight in aerial operations, with clear avenues for hardware upgrades and algorithmic enhancements.

Abstract

Detecting anomalies in hyperspectral image data, i.e. regions which are spectrally distinct from the image background, is a common task in hyperspectral imaging. Such regions may represent interesting objects to human operators, but obtaining results often requires post-processing of captured data, delaying insight. To address this limitation, we apply an anomaly detection algorithm to a visible and near-infrared (VNIR) push-broom hyperspectral image sensor in real time onboard a small uncrewed aerial system (UAS), exploring how UAS limitations affect the algorithm. As the generated anomaly information is much more concise than the raw hyperspectral data, it can feasibly be transmitted wirelessly. To detection, we couple an innovative and fast georectification algorithm that enables anomalous areas to be interactively investigated and characterized immediately by a human operator receiving the anomaly data at a ground station. Using these elements, we demonstrate a novel and complete end-to-end solution from data capture and preparation, through anomaly detection and transmission, to ground station display and interaction, all in real time and with relatively low cost components.

Figures (14)

• Figure 1: System data flow
• Figure 2: Anomaly test scene
• Figure 3: Detector Receiver Operating Characteristic (ROC) curves
• Figure 4: UAS and flight system (front on the right, two legs in the middle)
• Figure 5: Ground control station with radio and antennas mounted