Output details

Spectral unmixing with negative and superunity abundances for subpixel anomaly detection

<23> This paper proposes an innovative low false alarm methodology to determine anomalies in hyperspectral data. This research work is the outcome of a long-term collaboration between O. Duran and Prof. M. Petrou at Imperial College London, U.K. The outcome of this work, funded by MOD, allowed to identify real man made anomalies in natural backgrounds and distinguish them from rare natural objects. This technique was successfully disseminated in MOD annual conferences, has attracted a good number of citations and has potentially very significant implications for safety (e.g. antipersonnel mine detection).