Master of Science (M.S.), University of Dayton, 2018, Electro-Optics

Imaging through obscurants is a critical issue for lidars looking through clouds, or human tissue. Traditionally Spatial heterodyne imaging has been performed with a low-bandwidth laser source that exhibits good coherence length characteristics. One of the drawbacks of using a low-bandwidth source with long coherence length is that signal return from all objects within the coherence length of the source mix equally well on the camera imaging the system. Broadening the bandwidth of the source shortens the coherence length of the system. This thesis intends to show that through careful system design, spatial heterodyne imaging can be performed in the presence of a broadband source, allowing significantly improved imaging in the presence of obscurants such as clouds or human tissue. The method used will be phase modulating the source with a pseudo-random bit sequence and matching the optical path lengths of the signal and local oscillator branches of the system. By matching the path lengths for a pseudo-random coded source we can image objects at specific distances related to the modulation speed and code length, while isolating the power of signal return from objects at other distances as a factor of the autocorrelation coefficient of the code.

Committee: Paul McManamon Ph. D. (Advisor); Edward Watson Ph. D. (Committee Member); Partha Banerjee Ph. D. (Committee Member) Subjects: Electrical Engineering; Engineering; Optics; Remote Sensing

Master of Science (M.S.), University of Dayton, 2013, Electro-Optics

Two wavelength coherent imaging is a technique that offers several advantages over conventional coherent imaging. A significant advantage examined in this thesis is the ability to extract 3D target relief information from the phase contrast image at a known difference frequency. However, phase noise detracts from the accuracy that the target can be faithfully identified. We therefore describe a method for developing a relation of phase noise relative to the correlation of the image planes corresponding to each wavelength. Being able to predict the phase noise spectrum of a scene will help greatly in determining our ability to reconstruct the target relief. We examine the validity of previously derived equations, and extend them to a general case, which allows for the calculation of a correlation of a complex image field which has content from many spatial frequencies. The correlation coefficient can be used to generate a probability density function which represents the overall phase noise of the system relative to the spatial frequency content. For a simple target the spatial frequency content is based on a single tilt angle for the target. We discuss both computer-based modeling that is compared to an analytic equation, as well as an experimental spatial heterodyne verification of the model. We further extend our theory by building scenes with complex objects, discussing whether the derived equations hold for multi-faceted surfaces.

Committee: Joseph Haus (Advisor); David Rabb (Committee Member); Paul McManamon (Committee Member); Partha Banerjee (Committee Member) Subjects: Electrical Engineering; Optics