J. D. B. Nelson

Learn More
Target detection from hyperspectral imagery requires the fusion of information from hundreds of spectral bands. In this paper, we study such fusion in the context of hyperspectral image classification. Previous research applying kernel methods such as support vector machines (SVMs) to hyperspectral image classification has achieved performance competitive(More)
Hyperspectral imagery consists of hundreds of spectra or bands whose intensity is measured at various wavelength. Fusing the multiple spectral bands can provide more potential to differentiate between natural and man-made objects, and significantly improve the capability of target detection and classification. Spectral band or wavelength selection is one of(More)
—Sand ripples present a difficult challenge to current mine hunting approaches. We propose a robust and adaptive method that suppresses sand ripples prior to the detection stage. The method exploits a fractal model of the seabed and the connection between: dual-tree wavelets and local, directional fractal dimension; interscale energy ratios, scale invariant(More)
Semi-local Hurst estimation is considered for random fields where the regularity varies in a piecewise manner. The recently developed generalised lasso is exploited to propose a spatially regularised Hurst estimator. Dual-tree complex wavelets are used to formulate the usual log-spectrum regression problem and an interlaced penalty matrix is constructed to(More)
We here establish and exploit the result that 2D isotropic self-similar fields beget quasi-decorrelated wavelet coefficients and that the resulting localised log sample second moment statistic is asymptotically normal. This leads to the development of a semi-local scaling exponent estimation framework with optimally modified weights. Furthermore, recent(More)
—The detection and tracking of targets in aerial imagery of cluttered urban environments is addressed. Polar matching, using dual-tree complex wavelet transforms, is used as a shift and rotation invariant detector. A particle filter is employed to add robustness, especially in the event of target occlusion. We show that, together, these methods can robustly(More)
A recent dual-tree wavelet shrinkage method to suppress sand ripples in sonar imagery is extended with a Markov random field framework. Markov chain Monte Carlo sampling is used to estimate the posterior marginal ripple state in the wavelet domain. Ripple suppression is realised by multiplying the dual-tree wavelet coefficients by the conditional(More)
A generalised Lasso iteratively reweighted scheme is here introduced to perform spatially regularised Hurst estimation on semi-local, weakly self-similar processes. This is extended further to the robust, heavy-tailed case whereupon the generalised M-Lasso is proposed. The design successfully incorporates both a spatial derivative in the generalised Lasso(More)
Semi-local Hurst estimation is considered by incorporating a Markov random field model to constrain a wavelet-based pointwise Hurst estimator. This results in an estimator which is able to exploit the spatial regularities of a piecewise parametric varying Hurst parameter. The pointwise estimates are jointly inferred along with the parametric form of the(More)