Investigations on Complex Wave Retrieval Algorithms Using Optimized Compressive Fresnel Digital Holography and Its Applications In Information Processing

Abstract

Compressive Sensing (CS) is a paradigm shift in the signal sampling theory that newlinehelps in signal reconstruction from a finite number of measurements by exploiting the newlinesparsity of the signal. The CS can reconstruct signals and high-resolution images at a newlinesampling rate far less than Nyquist sampling criteria. Compressive digital holography newlineis a new approach in digital holography that allows reconstruction of the original object newlinewavefront from fewer hologram samples. The recent literature survey reveals that the newlineCS has been applied in many in-line and off-axis digital holographic schemes to reconstruct newlinethe object wavefield using single exposure digital holography by modelling the newlinenon-linear holography formulation as a linear process. This is a valid approximation newlineto achieve the linearity requirement for the CS implementation, which at the same time newlinecould affect the accuracy of whole information reconstruction, especially the phase information newlineof the object wavefield. The significance of digital holographic system lies newlinein its ability to reconstruct the phase (depth information) of the object wavefield along newlinewith the intensity information. Moreover, the aspect of phase reconstruction ability newlineof digital holography is not found discussed much in many of the compressive digital newlineholography approaches. newlineThe proposed thesis investigates the use of CS in single-shot digital holography to newlinehandle complex-valued object information using optimized compressive Fresnel digital newlineholography. Two major challenges involved in the compressive Fresnel digital holography newlineto reconstruct the whole object wavefront (i.e., both the intensity and phase of newlineobject wavefield) without losing the essential information are investigated: Primarily newlinethe selection of the most suitable linear sensing matrix operator, and secondly the best newlinesparsification operator to be used in CS framework. Compressive digital holography newlinehas been applied in the field of optical cryptosystems. The 3-D data storage requirement newlineand transmission of complex object information