Abstract
We have developed an architecture for spectral classification in spectral imaging. Adaptive kernels are designed using probabilistically-weighted principal component analysis. Simulation predicts orders-of-magnitude reduction in error rates, and a prototype system is currently under construction.
© 2012 Optical Society of America
PDF ArticleMore Like This
Matthew Dunlop-Gray, Phillip Poon, Esteban Vera, and Michael Gehm
FTh2B.5 Frontiers in Optics (FiO) 2014
M. J. Dunlop, P. A. Jansen, and M. E. Gehm
CMD3 Computational Optical Sensing and Imaging (COSI) 2011
M. J. Dunlop, P. K. Poon, D. R. Golish, E. M. Vera, and M. E. Gehm
CW2C.3 Computational Optical Sensing and Imaging (COSI) 2013