Honda Y.,NHK Science and Technical Research Laboratories |
Miyakawa K.,NHK Science and Technical Research Laboratories |
Kubota M.,NHK Science and Technical Research Laboratories |
Nanba M.,NHK Engineering System Inc. |
Egami N.,Kinki University
2013 26th International Vacuum Nanoelectronics Conference, IVNC 2013 | Year: 2013
A new active-matrix Spindt-type field emitter array for a FEA-HARP (field emitter array-high-gain avalanche rushing amorphous photoconductor) image sensor was designed. Simulation and experimental results showed that the new active-matrix drive circuit can shorten the response time and can potentially meet the HDTV standard. © 2013 IEEE.
Sugimoto T.,Science and Technology Research Laboratories |
Oode S.,Science and Technology Research Laboratories |
Nakayama Y.,Science and Technology Research Laboratories |
Okubo H.,NHK Engineering System Inc.
ITE Transactions on Media Technology and Applications | Year: 2015
A reproduction method for the 11 frontal channels of a 22.2 multichannel sound system (22.2 ch) over a direct-view display is investigated. NHK is planning to broadcast 8K Super Hi-Vision with 22.2 ch in a pilot broadcasting in 2016. A loudspeaker array frame has been developed for integration into the direct-view display as a home reproduction system and is used as a reproduction device of 22.2 ch. Localization of the frontal channels reproduced by the real loudspeakers is subjectively evaluated in terms of localization accuracy by estimating the difference from the intended localization for an 85-inch display. Three reproduction methods to synthesize the channels on the display are proposed and evaluated by comparison with a real loudspeaker placed at an exact channel position. Appropriate reproduction methods for the 11 frontal channels of 22.2 ch were decided from the results of a subjective evaluation. © 2015 by ITE Transactions on Media Technology and Applications (MTA).
Okuda M.,NHK Science and Technical Research Laboratories |
Okuda M.,Tokyo Institute of Technology |
Miyamoto Y.,NHK Science and Technical Research Laboratories |
Miyamoto Y.,NHK Engineering System Inc. |
And 4 more authors.
Journal of Applied Physics | Year: 2015
Current-driven magnetic domain walls in magnetic nanowires have attracted a great deal of interest in terms of both physical studies and engineering applications. The anomalous Hall effect measurement is widely used for detecting the magnetization direction of current-driven magnetic domains in a magnetic nanowire. However, the problem with this measurement is that the detection point for current-driven domain wall motion is fixed at only the installed sensing wire across the specimen nanowire. A potential solution is the magnetic domain scope method, whereby the distribution of the magnetic flux leaking from the specimen can be analyzed directly by contact-scanning a tunneling magnetoresistive field sensor on a sample. In this study, we fabricated specimen nanowires consisting of [Co (0.3)/Pd (1.2)]21/Ta(3) films (units in nm) with perpendicular magnetic anisotropy on Si substrates. A tunneling magnetoresistive sensor was placed on the nanowire surface and a predetermined current pulse was applied. Real-time detection of the current-driven magnetic domain motion was successful in that the resistance of the tunneling magnetoresistive sensor was changed with the magnetization direction beneath the sensor. This demonstrates that magnetic domain detection using a tunneling magnetoresistive sensor is effective for the direct analysis of micro magnetic domain motion. © 2015 AIP Publishing LLC.
Kobayashi A.,NHK Engineering System. Inc. |
Onoe K.,NHK Japan Broadcasting Coporation Science and Technology Research Laboratories |
Ichiki M.,NHK Japan Broadcasting Coporation Science and Technology Research Laboratories |
Sato S.,NHK Japan Broadcasting Coporation Science and Technology Research Laboratories
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | Year: 2016
This paper compares unsupervised sequence training techniques for deep neural networks (DNN) for broadcast transcriptions. Recent progress in digital archiving of broadcast content has made it easier to access large amounts of speech data. Such archived data will be helpful for acoustic/language modeling in live-broadcast captioning based on automatic speech recognition (ASR). In Japanese broadcasts, however, archived programs, e.g., sports news, do not always have closed-captions used typically as references. Thus, unsupervised adaptation techniques are needed for performance improvements even when a DNN is used as an acoustic model. In this paper, we compared three unsupervised sequence adaptation techniques: maximum a posteriori (MAP), entropy minimization, and Bayes risk minimization. Experimental results for transcribing sports news programs showed that the best ASR performance is brought about by Bayes risk minimization which reflects information as to expected errors, while comparable results are obtained with MAP, the simplest way of unsupervised sequence adaptation. © 2016 IEEE.
Sugimoto T.,NHK Science and Technical Research Laboratories |
Ando A.,NHK Science and Technical Research Laboratories |
Ono K.,NHK Engineering System Inc. |
Morita Y.,Foster Electrical Co. |
And 3 more authors.
Journal of the Acoustical Society of America | Year: 2013
A lightweight push-pull acoustic transducer using dielectric elastomer films was proposed for use in advanced audio systems in homes. The push-pull structure consists of two dielectric elastomer films developed to serve as an electroactive polymer. The transducer utilizes the change in the surface area of the dielectric elastomer film, induced by an electric-field-induced change in the thickness, for sound generation. The resonance frequency of the transducer was derived from modeling the push-pull configuration to estimate the lower limit of the frequency range. Measurement results presented an advantage of push-pull driving in the suppression of harmonic distortion. © 2013 Acoustical Society of America.
Nishimoto Y.,NHK Japan Broadcasting Corporation |
Kawakita H.,NHK Japan Broadcasting Corporation |
Yamamura C.,NHK Japan Broadcasting Corporation |
Nakagawa T.,NHK Japan Broadcasting Corporation |
Inoue T.,NHK Engineering System Inc.
Digest of Technical Papers - IEEE International Conference on Consumer Electronics | Year: 2014
This paper presents is a downloadable conditional access system (CAS) for advanced satellite broadcasting with 4K/8K UHDTV. This CAS can perform secure access control and has security functions such as key data processing and encryption processing that can be securely updated through broadcasting or communications functions. A prototype TV receiver with the downloadable CAS was developed in order to evaluate feasibility of the proposed system. © 2014 IEEE.
Yasue T.,NHK Science and Technical Research Laboratories |
Kitamura K.,NHK Science and Technical Research Laboratories |
Watabe T.,Brookman Technology Inc. |
Watabe T.,University of Shizuoka |
And 8 more authors.
IEEE Transactions on Electron Devices | Year: 2016
A 1.7-in, 33-Mpixel, 120-frames/s, 14-bit CMOS image sensor has been developed. The 7936 (H) × 4412 (V) pixel CMOS image sensor, which uses 14-b depletion-mode MOS (DMOS) capacitor-based two-stage cyclic A/D converters (ADCs) and 64 parallel scalable low-voltage signaling output ports, operates at a data rate of 63.8 Gb/s. DMOS capacitors have a high capacitance density, but it is difficult to achieve high bit resolutions in ADCs with these capacitors because their capacitance depends on the applied voltage. Column-parallel two-stage cyclic ADCs overcome this difficulty using a split-sampling DMOS capacitors architecture. The two-stage cyclic ADC with the DMOS capacitors at a 6.4-μm column pitch exhibited a differential nonlinearity of 0.95/-0.80 least significant bit (LSB); the integral nonlinearity was 2.57/-28.27 LSB at a 14-b resolution. The CMOS image sensor implemented with a 90-/65-nm technology exhibited a sensitivity of 5.22 V/lx .s and a random noise of 3.6 e- rms with a gain of 3.3 at 120 frames/s while dissipating 3.2 W. © 2015 IEEE.
Ono K.,NHK Engineering System Inc. |
Nishiguchi T.,NHK Science and Technical Research Laboratories |
Matsui K.,NHK Science and Technical Research Laboratories |
Hamasaki K.,NHK Science and Technical Research Laboratories
135th Audio Engineering Society Convention 2013 | Year: 2013
NHK has been developing a portable spherical microphone for the simultaneous recording of 22.2ch multichannel audio. The microphone is 45 cm in diameter and has acoustic baffles that partition the sphere into angular segments, in each of which an omnidirectional microphone element is mounted. Owing to the effect of the baffles, each segment has narrow angle directivity and a constant beam width in higher frequencies above 6 kHz. The directivity becomes wider as frequency decreases and that it becomes almost omnidirectional below 500 Hz. The authors also developed a signal processing method, that improves the directivity below 800 Hz.
Arai J.,NHK Japan Broadcasting Corporation |
Yamashita T.,NHK Japan Broadcasting Corporation |
Hiura H.,NHK Japan Broadcasting Corporation |
Miura M.,NHK Japan Broadcasting Corporation |
And 3 more authors.
Proceedings of SPIE - The International Society for Optical Engineering | Year: 2015
A compact integral three-dimensional (3D) imaging device for capturing high resolution 3D images has been developed that positions the lens array and image sensor close together. Unlike the conventional scheme, where a camera lens is used to project the elemental images generated by the lens array onto the image sensor, the developed device combines the lens array and image sensor into one unit and makes no use of a camera lens. In order to capture high resolution 3D images, a high resolution imaging sensor and a lens array composed of many elemental lenses are required, and in an experimental setup, a CMOS image sensor circuit patterned with multiple exposures and a multiple lens array were used. Two types of optics were implemented for controlling the depth of 3D images. The first type was a convex lens that is suitable for compressing a relatively large object space, and the second was an afocal lens array that is suitable for capturing a relatively small object space without depth distortion. The objects captured with the imaging device and depth control optics were reconstructed as 3D images by using display equipment consisting of a liquid crystal panel and a lens array. The reconstructed images were found to have appropriate motion parallax. © 2015 SPIE.
Kanazawa M.,NHK Engineering System Inc.
Journal of the Society for Information Display | Year: 2013
Monochrome reflective-type displays are widely used for portable reading applications such as electric papers because this type does not need a back light unit and can be used outdoors for a long time. Color reflective-type displays without back light units are desired to expand the market further. The current color reproduction is based on three sub-pixel red, green and blue (RGB) methods, and when used in reflective type, its luminance is reduced to a third of that of monochrome type. Adding a white sub-pixel to the current method can improve the luminance, making the sub-pixel number four. However, in the case of a high resolution display with a four-sub-pixel method, the pixel structure is complex, and the luminance improvement may be limited. Instead of increasing the sub-pixel number, two sub-pixel methods are investigated. These methods can improve luminance with limited color gamut. The performances are compared with those of other methods quantitatively. © Copyright 2013 Society for Information Display.