Abstract
A novel anomaly detection method for hyperspectral images (HSIs) is proposed based on anisotropic spatial-spectral total variation and sparse constraint. HSIs are assumed to be not only smooth in spectral dimension but also piecewise smooth in spatial dimension. The proposed method adopts the anisotropic spatial-spectral total variation model which combines 2D spatial total variation and 1D spectral variation to explore the spatial-spectral smooth property of HSIs. Meanwhile, the sparse property of anomalies is exploited for its low probability in the image. To utilize both the spatial and spectral information of HSIs, we preserve the original cubic form of HSIs and divide the HSIs into three 3D arrays, each representing the background, the anomaly, and the noise respectively. By using anisotropic spatial-spectral total variation regularization on the background component and sparse constraint on the anomaly component, this anomaly detection problem has therefore been formulated as a constraint optimization problem whose solution has been derived by alternately using Split Bregman Method and Go Decomposition (GoDec) Method. Experimental results on hyperspectral datasets illustrate that our proposed method has a better detection performance than state-of-the-art hyperspectral anomaly detection methods.
Highlights
Hyperspectral sensors can collect a spectral vector with hundreds or thousands of elements from each pixel in a given scene(Bioucas-Dias et al, 2013), and the data product, which is generally termed hyperspectral image (HSI), is a three-dimensional array or “cube” with the width and length of the array corresponding to spatial dimension and the spectrum of each point as the third dimension(Stein et al, 2002)
After removing the bands that correspond to the water absorption regions, low signal-to-noise ratio (SNR), and bad bands, 189 available bands of the data were retained in our experiments
ASSTVSC has been proposed from a novel perspective to employ total variation model for hyperspectral anomaly detection
Summary
Hyperspectral sensors can collect a spectral vector with hundreds or thousands of elements from each pixel in a given scene(Bioucas-Dias et al, 2013), and the data product, which is generally termed hyperspectral image (HSI), is a three-dimensional array or “cube” with the width and length of the array corresponding to spatial dimension and the spectrum of each point as the third dimension(Stein et al, 2002). In addition to aforementioned methods, recently, lowrank and sparse matrix decomposition-based methods for HSIs AD have gained much attention This idea comes from that the background has the low-rank property due to the high correlation of spectral information (Zhang et al, 2014) and the anomalies are sparse for its low probability (Cui et al , 2014). Considering the ubiquitous mixed pixels in HSIs, Liu (Liu et al, 2010) assumes that the low-rank matrix representing the background should lie in multiple subspaces instead of a single subspace proposed in RPCA model. Combing with the above analysis, a novel anomaly detection method based on anisotropic spatial-spectral total variation and sparse constraint (ASSTVSC) is proposed with the assumption that the background has the spatial-spectral smooth property, the anomaly is sparse and the noise is independent and identically distributed Gaussian noise
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
