Hyperspectral Imaging of Chemical Composition

Jerald C Martin, Purdue University


In this work, two different types of imaging are explored, one using fluorescence and another using X-rays. Hyperspectral imaging is a method of imaging materials that typically has three dimensions of information: x, y, and λ (wavelength). These measurements are useful for broad classes of imaging applications, including astronomy, geology, and medicine. However, the current methods used for hyperspectral imaging have significant drawbacks, including 1/f noise and complexity. Therefore, new methods have been recently developed to address these issues, such as new snapshot imaging systems. These systems, however, still often have high complexity, are expensive, and have limited applicability outside their initial application. Therefore, a new system has been developed that can be integrated into most beam scanning microscopes that is both fast (can produce up to 17 frames per second) and can classify components in a complex mixture. This includes being able to identify regions in a live biological sample. In addition, a powder X-ray diffraction scheme is shown to reduce time and amount of material needed for measurement. This technique disentangles different orientations within a limited sample to increase confidence in the measurements, as with a limited amount of material it is no longer an average of all possible orientations.




Simpson, Purdue University.

Subject Area


Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server