Output details

Robust endmember extraction in the presence of anomalies

<23> This paper proposed a method to identify endmembers in a robust way, separating them from outliers present in the scene. The latter are often confused by other methods as endmember spectra, whereas in practical applications, they are treated as anomalies or targets.The outcomes of this technique may be applied to other fields of signal and image processing. In particular, this well received paper has led to industrial and academic collaborations with Arnia Ldt and the Polytechnic Institute of Automation in Poland, respectively, to address the decline in honeybee populations by identifying the sound from bees’ sources.