We controlled the fine-pitch polymer aggregation in liquid crystals (LCs) two-dimensionally via patterned ultraviolet (UV) irradiation, where a photomask was used to precisely control the light distribution. The combination of high-intensity UV light and a polymerization inhibitor suppressed polymerization in the masked area, and improved the uniformity of LC alignment between fine-pitch polymer structures. We were able to electrically control the diffraction patterns of polymer-dispersed LCs.

We developed a high-speed, electrically tunable multi-bandpass filter using a polymer-stabilized blue-phase liquid crystal etalon and a narrowband multi-bandpass interference filter, for use in real-time multi-spectral imaging systems. We show theoretically that the filter exhibits high-level transmittance (≥ 80%) by using an Ag film as the reflective film used in the etalon. We confirmed that the filter allows for high-speed switching of the transmitted wavelength; the response time is about 1 ms. The filter enables high-transmittance, high-speed spectroscopy. Therefore, high-sensitivity real-time multi-spectral imaging at a frame rate of 60 fps or more can be expected in the future.

We quantitatively investigated the effect of two non-uniform phase distributions in parallel-aligned liquid crystal on silicon spatial light modulators on the image quality of a holographic image through numerical simulation. We conclude that, in the presence of a locally non-uniform phase distribution due to non-uniform liquid crystal (LC) alignment within a pixel, the diffraction efficiency of the first-order diffracted light decreases, and the brightness and uniformity of intensity of the image degrade. This is because the phase distribution near the boundary with the adjacent pixel changes from a rectangular to a sinusoidal wave shape. When there is a globally non-uniform phase distribution due to uneven thickness of the LC layer across the entire panel, the resolution characteristics and noise are affected. This is because the first-order diffracted light deviating from the original emission angle is present and the condensing point of the first-order diffracted light expands.

For realization of phase shifting using a liquid crystal (LC) that can continuously control the phase of microwaves, we evaluated the relationship between the molecular structure of the LC, which assumes a nematic phase at room temperature, and the dielectric loss in microwave frequency. The results indicated that the LC has a rigid molecular structure, has fluorine as a polar group, and is in a low temperature, leading to lower dielectric loss. From these results, we considered that dielectric loss can be reduced by suppression of the thermal vibration of LC molecules under microwave exposure.

High performance IGZO TFTs with top gate structure were developed for an automotive OLED display backplane. Fabrication processes are optimized by balancing oxygen and hydrogen contents with μ-PCD method. The mobility of the IGZO TFTs reaches as high as 32 cm²/Vs with enhanced threshold voltages. We have checked the TFTs reliability under the positive bias temperature (PBT), negative bias temperature (NBT) and negative bias temperature illumination (NBTI) stress tests. As the IGZO TFTs shows slight changes of threshold voltage (V_th) within ±0.5V under PBT and NBT and even after NBTI stress tests, there is no critical deterioration. We expect these high mobility IGZO TFTs are stable enough to be used for OLED or other self-luminous displays. We have also demonstrated a prototype 12.3" OLED module for automotive applications. The prototype flexible display showed an excellent brightness uniformity even after bending.

This paper proposes a super-multiview autostereoscopic display with a full HD resolution. A super-multiview display, where multiple views are generated around the eyes to stimulate focal accommodation, requires generation of images for many viewpoints. When we apply time-division multiplexing, the number of views can be increased without losing spatial resolutions, while flickers stand out as the number of time-division increases. To attain more views with the same refresh rate, we set two LCD panels so that they may face the opposite directions. In this way, the order of color filter is reversed and the light rays of different colors are directed to different orientations. Each color creates a different directional light to achieve three fold views. To extend the viewing zone in the depth direction, we introduce adaptive time-division, where quadruplexing is applied when the viewer is farthest, quintuplexing is applied when the viewer is in the middle, and sextuplexing is applied when the viewer is nearest. Expansion of viewing zone and the effect of focal induction are confirmed by the experiments using a prototype system based on the proposed method.

We propose a new type of volume holographic waveguide using multiplex recording for head mounted displays called the volume-holographic multiplexed-mirror (VHM) waveguide. The VHM waveguide is based on a photopolymer with a thickness of several hundred micrometers that has been developed for holographic data storage. We investigated if the VHM can exhibit high performance when combined with broad wavelength light sources. We prototyped a head mounted display unit with the VHM waveguide and evaluated the transmittance, luminance efficiency, and potential field of view. We confirmed clear transparency of 87%, high luminance efficiency of 3850 cd/m2/lm suitable for outside use, and wide potential field of view of 35 degrees (horizontal) and 36 degrees (vertical). The VHM waveguide could pave the way for new high-performance head mounted displays for augmented reality.

In this study, we conducted an experiment using a mobile augmented reality (AR) system that allows direct manipulation of a virtual object with the user's hand and that presents material appearance of the object. The system presents user-perspective images on the mobile display, which enables natural interaction with virtual objects. We compared the methods with user-perspective and camera-perspective presentation using an object manipulation task for material perception. The result showed that the answer time in the task became significantly shorter with user-perspective presentation, which indicates that user-perspective presentation makes manipulation of virtual objects in mobile AR easier.

A whole slide imaging (WSI) scanner scans pathological-specimens to produce digital images for monitor-based diagnosis and analysis. However, the image quality is sometimes insufficient due to focus-error or noise, in which case the slide needs to be rescanned. In previous work, a referenceless quality evaluation technique was proposed, but some artifacts (i.e. tissue-fold, air-bubble) were detected as false positives. Those artifacts need to be ignored in determining whether rescanning is necessary or not, because they are not caused in the scanning but slide preparation stage. This paper proposes a method for a more practical system to assess WSI quality by distinguishing the origins of quality degradation; the focus-error or noise caused by the scanner and the artifact occurred in the slide preparation. In the method, a support vector machine detects artifacts first, and then quality is evaluated excluding artifact regions. The effectiveness of the proposed system has been experimentally demonstrated.

In this paper, we propose a multiple human tracking method with alternately updating trajectories and mult iframe action features (MHT-MAF). Even though occlusion or motion blur occurs due to the sudden movement of the drone, ID switches are prevented by the stable MAF. In the experiments, we verified the effectiveness of the proposed method using the Okutama-Action dataset. Our code is available online (https://github.com/hitottiez/mht-paf).

A prototyped 8K stacked CMOS image sensor overlaid with a crystalline-selenium-based avalanche-multiplication layer, in which holes are used as traveling carriers in the film, was fabricated. Analysis of energy-band diagrams through the film to the n-type floating-diffusion region revealed that (i) large spot noise in the captured image could be suppressed and (ii) the high voltage required for avalanche multiplication could be applied to the film by using holes as carriers even when defects existed in the film. According to the results of experiments, no large spot noise occurred when the voltage applied to the film was +5 V. Additionally, the photoelectric-conversion current was increased by 1.4 times compared to the saturation-signal level when the applied voltage was +21.6 V. These results confirm charge multiplication in a crystalline-selenium-based stacked CMOS image sensor.