Hyperspectral Data Processing: Algorithm Design and Analysis

33.8 Applications

Despite that only a few applications are presented in Category C in this book many other applications such as agriculture food quality and safety inspection, homeland security, chemical/biological sensing, and medical imaging, have emerged as major players of hyperspectral imaging sensors in various SPIE conferences, International Symposium on Spectral Sensing Research (ISSSR) and IEEE symposia, conferences, and workshops. The applications presented in this book are only of particular interest to the author's preference. The two applications considered in Chapter 30 are subpixel target size estimation and concealed target detection, both of which pose difficulties for conventional spatial domain-based techniques since targets of interest in these applications cannot be visualized by inspection and can only be resolved by their spectral not spatial information. As noted in Sections 1.2 and 1.3, HyperSpectral imaging (HSI) is not a direct extension to MultiSpectral imaging (MSI). As a result, techniques developed for hyperspectral imaging are not necessarily applicable to multispectral imagery. To make hyperspectral imaging techniques also effective in multispectral image analysis, Chapter 31 explores differences between HSI and MSI and further develops two approaches to nonlinear dimensionality extension, band dimensionality expansion (BDE), and kernelization to expand the ability of HSI in dealing with multispectral imagery. Chapter 32 presents one of promising ...