Hyperspectral anomaly detector based on variable number of linear predictors

An important application in remote sensing using hyper-spectral imaging system is the detection of anomalies in a large background. An anomaly detector for hyperspectral imagery is developed by partialling out the effect of the clutter subspace by predicting the background using a linear combination of the clutter subspace. The coefficients of the linear combination are chosen to maximize a criterion based on squared correlation. The dimension of the clutter subspace for each spectral component of the background can vary from one spectral component to another. The experimental results from a hyperspectral data cube show that the anomaly detector has a better performance than the SSRX detector.