Mixed reality or augmented/virtual reality is a bridge technology between real life and virtual life. Mixed reality is getting a spotlight because it can interact with real life, unlike conventional 3D display which delivers information on one-way direction. There are several candidates as 3D displays such as a multiview display, integral imaging display, volumetric display, and holographic display. The holographic display is an ultimate 3D display theoretically which produces perfect virtual environment consistent with real life. However, spatial frequency and bandwidth of spatial light modulator restrict the resolution and eld of view of a hologram. Besides, it takes a significant amount of time to calculate a hologram pattern. Although a Maxwellian-view display cannot provide monocular depth information, it is widely used to near-eye display because it solves vergence-accommodation mismatch and eld of view.
This dissertation covers the integration of the Maxwellian-view and holographic display to solve 1. resolution & calculation time, 2. system form factor & speckles among various problems with the holographic display. Resolution of the hologram is proportional to the number of pixels of the spatial light modulator, and it is inversely proportional to a distance from the spatial light modulator to the hologram. Hence, resolution of hologram becomes decreasing which corresponds to airy disk size when voxels formed far from the spatial light modulator. Therefore, the number of voxels and sharpness of hologram always have a limitation. In general, the spatial light modulator with small pixel pitch and a large number of pixels is a fundamental solution, but it is not easy because of the limitations of current semiconductor technology. Besides, the hologram pattern (Interference pattern) calculation time increases in proportion to the number of voxels to form in the air.
In this dissertation, depth layer method was used which has a relatively fast calculation time, but it still took lots of time to render entire virtual objects, and it could not produce complex virtual objects due to the resolution limit. Therefore, in order to enhance the resolution of the hologram and reduce calculation time, the Maxwellian-view display is combined with the holographic display by time-division multiplexing. Hence, resolution and calculation time of hologram is highly improved by processing only small portion corresponding to fovea with the holographic display. An artificial depth cue, which is the same as the optical characteristics of the eye, is applied to a remaining region and the Maxwellian-view display represents it.
For near-eye display, simple structure and light weight are essential. Therefore, a switchable holographic optical element suggested to simplifying the system structure and form factor instead of a beam splitter and eyepiece lens. Characteristics of the holographic film analyzed with Kogelnik's coupled-wave theory, and an optimized structure drew. Then, multiple optical systems in a single holographic film are verified ed by controlling the reference wave with a liquid lens. Besides, holographic displays widely use lasers which have high spatial and temporal coherence. However, not only it is harmful to the human body, but it reduces the hologram quality because of high interference that strengthens hologram speckle. Therefore, spatial coherence and speckle variations verified depending on LED and pinhole size and also confirmed that holograms implemented.
Nevertheless, while there are issues to be addressed such as the size of the viewing window and color, the proposed holographic near-eye display is sufficiently proven as an AR/VR near-eye display.