Comprehensive Performance Analysis of Objective Quality Metrics for Digital Holography

摘要: Objective quality assessment of digital holograms has proven to be a challenging task. While prediction perceptual the recorded 3D content from holographic wavefield is an open problem; after rendering, requires time-consuming rendering step and multitude possible viewports. In this research, we use 96 Fourier recently released HoloDB database evaluate performance well-known state-of-the-art image metrics on holograms. We compare reference with their distorted versions: (i) before real imaginary parts quantized complex-wavefield, (ii) converting Fresnel holograms, (iii) amplitude reconstructed data, (iv) subsequently removing speckle noise using Wiener filter. For every experimental track, metric predictions are compared Mean Opinion Scores (MOS) gathered 2D screen, light field display display. Additionally, statistical analysis results discussion presented. The tests demonstrate that while for each test track few present highly correlated multiple sets available MOS, none them demonstrates consistently high-performance across all four test-tracks.