A wavelet transform and linear discriminant analysis were applied to spectral traces of materials to improve material fractional abundance estimation by reducing material class sizes and increasing class separations. This approach is demonstrated using a variety of true material traces that were used to generate simulated measurements.
© 2004 Optical Society of America
K. Hamada, S. M. Giffin, and C. L. Matson, "Feature extraction using wavelets to improve the estimation of material fractional abundances," in Frontiers in Optics 2004/Laser Science XXII/Diffractive Optics and Micro-Optics/Optical Fabrication and Testing, OSA Technical Digest Series (Optical Society of America, 2004), paper FTuB5.
References are not available for this paper.