To find the minimum resolution required for visual realness, we performed subjective assessments to examine the realness of images at various resolutions. A paired-comparison experiment quantified the realness of images by comparing images of different resolutions or by comparing images with the real object. Observers were asked to choose the image that appeared to be most like the real object. The results indicated that the realness of an image increased with image resolution, reaching a plateau at about 80-100 pixels per degree.

A wide field-of-view visual stimulus with homogeneous motion gives us an alternative perception of vection and object-motion. Body sway caused by vection has been used as an objective index of the sensation of presence, or the sensation of "being there." We used a near infrared spectrometer in an attempt to detect changes in blood flow in the brain when a human perceives vection as an objective index of the sensation of presence.

Research on the perception of presence and spatial cognition is being carried out in partnership with Waseda University.

We measured the average minimum audible field (MAF) for people ranging in age from their twenties to their eighties and investigated ways to adapt the speech in broadcast programs to aural characteristics of elderly listeners. The method adjusts the speech signal energy with reference to the partitioned frequency band below the average MAF. Subjective listening tests confirmed that elderly listeners prefer the adjusted sounds.

Large vocabulary continuous speech recognition is expected to be useful in many broadcasting applications, such as closed captioning, as its capability becomes more human-like. Our recent progress on speech recognition including acoustic models based on the auditory mechanism, language models utilizing semantic information, and methods to assess and learn from word error tendencies, have resulted in higher word recognition accuracy.

Polymer thin-film transistors (TFTs), based on semiconducting polymers with a liquid-crystalline nature, have potential uses in flexible active-matrix displays. We obtained highly crystalline and oriented semiconducting polymer films by treating the surface of the substrate and annealing the films. The resulting polymer TFTs had a high mobility and a high current on/off ratio. They also had very good stability in air.

Our unique catalyst technology has been used to fabricate a super-fine pitch field emission display employing graphite nanofiber (GNF) as an efficient electron emission material. GNF is a carbon nano-material consisting of nano-meter sized fine fibers, and it is easy to use to fabricate a fine structure for a large screen. Our FED is a unique means of creating from compact to over 100 inch Super Hi-Vision displays with low power consumption for home use.

We fabricated efficient phosphors of ternary compounds based on a new luminescence mechanism. This research will lead to brighter and more efficient phosphors. The phosphors emit bright luminescence through a mechanism wherein rare-earth ions in the phosphor are excited by hot hole carriers, rather than the hitherto known hot electron carriers. The purity of the phosphors is the highest in the world thanks to our original fabrication technique.

We fabricated novel light modulation devices with submicron lateral dimensions and confirmed their basic operations. Incident light is reflected by the ferromagnetic layer of the device with the rotation of the polarization plane, the rotation angle of which varies along with the magnetization direction (the magneto-optical Kerr effect). The magnetization direction is controlled by spin polarized current (spintransfer switching).

This Research is being carried out in partnership with the Institute for Solid State Physics, University of Tokyo.

We have been studying holographic data storage as a potential mass storage system for Super Hi-Vision video materials. An experimental holographic data storage system has been developed. This system combines a unique method of phase compensation and a speckle multiplexing method using a random phase mask. We stored highly compressed moving pictures by using 200 multiplexed holograms, and confirmed that the moving pictures could be reproduced correctly.

We investigated an optical data reading device for videotape that might enable high-speed data copying from videotape to next-generation recording media. The combined use of magnetic garnet film and a laser beam would make it possible to collectively read the multiple recorded tracks on videotape. This presentation shows the magnetic garnet film data pattern transferred from videotape, and a primitive reading experiment.

This Research is being jointly conducted with the Shizuoka Institute of Science and Technology.
The magnetic garnet film is being developed in partnership with FDK Corporation.