Shot Noise and Fixed-Pattern Noise Effects on Digital Hologram Reconstruction

Abstract

Digital holography allows registering and reconstructing information about 3D objects and 3D scenes. There are several limitations related to reconstructed image quality: speckle noise, twin images, zero order, light (shot noise) and dark temporal noise, camera fixed-pattern noise (spatial noise), dynamic range, and quantization noise. These restrictions determine the maximum possible signal-to-noise ratio (SNR) of digital holograms and reconstructed images. This work is concerned with studying the effect of a digital camera's main noise components on hologram reconstruction. Analytical equations for estimation of SNR of reconstructed amplitude image were obtained. They relate to a number of values: shot noise, dark temporal noise, fixed-pattern noise, camera's dynamic range, quantization noise, reference and object beam intensities, and the ratio between the object area and the entire reconstructed field. The resulting equation was tested experimentally using digital holograms of diffusely scattering objects. Characteristics of different types of CCD and CMOS cameras were used: digital single-lens reflex (DSLR), scientific, industrial, and video surveillance cameras. The effect of the camera noise on the phase image reconstruction was also estimated. The resulting equations make it possible to estimate the camera noise's effect on the reconstructed images before the experiments were conducted. As a result, the illumination value, exposure time, object size, and other experimental conditions can be pre-selected to reduce the noise in holograms used in metrological and infrared applications and optical encryption.