Chemical imaging and hyperspectral imaging modalities, which are non-destructive and non-invasive, are gaining a lot of interest in industrial and research fields. In particular, Raman spectroscopy and Raman hyperspectral imaging are rapidly becoming analytical methods of choice. High resolution and high fidelity Raman Imaging is challenging, however, due to the performance limitations of commercially available electronically tunable filters. The emphasis of my research has been to develop high resolution spectral filters and Raman imaging system for bio-materials. Raman scattering is commonly used for the nondestructive identification of minerals and polymers, as well as many other materials, and it is suitable for characterizing the degree of crystallinity, which can impart material properties such as piezoelectricity. Another type of inelastic scattering, Brillouin scattering, provides an indirect measure of sample elasticity and results from the interaction of light with time dependent optical density (refractive index) variations imparted by acoustic phonons. Achieving the necessary spectral resolution (bandwidth“1nm) to separate Rayleigh scattered light and Brillouin scattered light is challenging, and conventional grating monochromators are often replaced with a cascaded series of Fabry-Perot filters. Developing electronically tunable filters for Brillouin scattering is extremely challenging. A more recent technology, the virtually imaged phased array (VIPA), is capable of narrow passbands and high angular dispersion. As a part of this work generalized dispersion law for the VIPA, employing both anisotropic and isotropic dielectrics has been developed. Low light applications like Raman and Brillouin imaging techniques can take advantage of spatial multiplexing strategies in which the intensity of light from many pixel locations on the sample are detected simultaneously. Cyclic simplex matrices have been employed as encoding masks for binary state multiplexing applications and are employed in this work as part of the development of a multiplexed Hadamard transform Raman imaging system. The Hadamard transform Raman instrument utilizes an electro mechanically controllable digital micro-mirror device (DMD) as a spatial modulator. The result is a significant improvement in the signal-to-noise ratio of the Raman hyperspectral data which can enable faster acquisition times. In addition, correction methods have been developed to completely remove or reduce errors caused by photobleaching and stray cosmic radiation. In parallel with this work, novel signal processing and multivariate analyses have been investigated for the particularly challenging application of the Raman modality to biological samples. In particular, Raman markers have been identified using these methods to differentiate normal retinal tissue from glaucomatous tissue in mice. Glaucoma is a major neurodegenerative disease that often leads to irreversible vision loss. Its early detection relies on new analytical methods, such as Raman spectroscopy and Raman imaging that can reveal early biochemical changes associated with its onset.