Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a generalized machine learning ensemble framework that jointly models NASA Black Marble M-band and Day/Night Band (DNB) signals to detect nighttime anomalies, overcoming limitations of thermal-band-only approaches. The framework iteratively builds spatio-temporally robust detectors that scale across diverse regions, seasons, and anomaly classes, improving true detection rates while reducing false positives. This approach demonstrates strong generalizability for applications in natural hazard monitoring and energy extraction.
Nighttime anomaly detection is significantly enhanced by fusing M-band and DNB satellite signals, achieving higher accuracy and broader applicability than thermal-only methods.
Nighttime lights from NASA's Black Marble product suite capture thermal and light emission signals from anomalous events including fires, volcanic eruptions, and gas flaring. Existing detection approaches rely primarily on thermal bands, limiting sensitivity to weaker signals. We propose a novel machine learning framework that jointly models Black Marble M-band and Day/Night Band (DNB) signals to derive a generalized, spatio-temporally robust ensemble of anomaly detectors. The framework iteratively builds detectors that scale across regions, seasons, anomaly classes, and extends over land and ocean. Detection sets at varying confidence levels are derived based on relevant bands and detector agreement. The approach improves true detection rate while reducing spurious detections and results demonstrate strong generalizability with applications in natural hazard monitoring and energy extraction.