Ga_2O_3 thin films were deposited on quartz substrates using Eu-doped GaN by Electron Beam evaporation, then they were annealed at several temperatures for 1 hour in Ar+O_2 flows. It was found that the as-deposited film consists of almost fraction of Gallium metal, and the Ga_2O_3 film is formed by the annealing.

抄録全体を表示

Organic EL devices were fabricated on p- Si substrate for the purpose of integrating organic thin-film EL devices on MOSFET. It was found that a buffer layer is necessary to improve the sticking properties of the organic layer to the Si substrate. EL from Alq_3 emitting layer could be observed when ITO was used buffer layer..

抄録全体を表示

Color image can be classified into different object areas with clustered color distribution. After the classification process, each object area is characterized by Principal Component Analysis. The a* and b* are predicted by the projection of L* onto chromatic plane using the eigen vectors of covariance matrix and the mean vectors. Finally, the full color image is reproducted by combining the luminance with (a*,b*).

抄録全体を表示

We previously proposed a method for designing predictors which can minimize a coding rate of prediction errors in lossless coding of still images. In this paper, we extend the method to lossless coding for volume data. Simulation results indicate that coding rates of the proposed lossless coding scheme are 0.02〜0.07 bit/voxel lower than those of the scheme using predictors designed on the MSE criterion.

抄録全体を表示

画像を関数表現するベクトル符号化方式は、拡大縮小に耐性があるだけではなく、編集処理を繰り返しても劣化がないため情報流通にも適している。さらに色の境界が明快な画像に適用した場合、符号量圧縮にもなるので文字や地図等では盛んに使用されている。しかし、自然画への適用は逆に符号量が膨大になることからこれまで十分な研究開発は行われていない。このような背景から、本稿では自然画に対して境界輪郭線を輝度値を高さとして等高線モデルとしてとらえ、類似変化を利用することにより符号量圧縮の実現を目指す。

抄録全体を表示

We previously proposed a motion compensation method which utilized the same motion vectors for both texture and shape information of arbitrarly-shaped VOP. In this paper, we optimize motion vectors and block sized iteratively to improve the eggiciency of warping prediction. Simulation results indicate that the proposed method is superior to the conventional method in terms of SNR of predicted images at the same bit rate.

抄録全体を表示

An object scalable coding scheme is proposed for the purpose of realization of the scalable coding which is based on the priority of each object in the picture. The features on the scheme include that the bit stream is comparible with SNR scalability, and that the SN ratio of decoded upper layer pictures can reach that in SNR scalable coding.

抄録全体を表示

In this paper, we study video format conversion method from MPEG-2 to MPEG-1. We have performed several transcoding experiments using direct motion vector conversion from MPEG-2 to MPEG-1, including a new proposal with simplified calculation. By adopting these vectors, MPEG-2 video is transcoded to MPEG-1 video without motion estimation. As a result, processing time for MPEG-1 re-encoding ca be reduced by less than half.

抄録全体を表示

In this paper, we proposed the video-transmission control method noted frame structure to protect video quality. This proposal method applies error control (retransmission) to referred frame not to spread the error. We evaluated the method by simulation, and proved the improvement of video quality.

抄録全体を表示

In this paper, we propose a video transmission and a image interpolation method taken advantage of frame reference structure on MPEG-2, which realize high-quality video transmission over ATM network. Experimental results show that our method realizes higher image quality than image quality of traditional method independently of traffic condition.

抄録全体を表示

This paper studies a coding control method for software real-time video communication. Since CPU performance is greatly improved recently, video codec system can be implemented by software. It is expected that video codec software is widely spread and installed on many kinds of computers. In case that performance difference between terminals is so large, the PC with lower performance may have some trouble due to bitstream encoded with high frame rate by the PC with higher performance. In this paper, a coding control method considering computer's performance is studied.

抄録全体を表示

This paper proposes a real-time MPEG-2 video software encoder on parallel computer systems and evaluates its performance. This parallel software encoder consists of a small number of high-speed encoding elements and a frame-synchronization controller with I/O. This encoder provides sufficient performance for real-time full NTSC-size encoding on a dual-PC.

抄録全体を表示

This paper describes an MPEG-2 real-time encoding TypeII PC-card system for mobile applications. The PC-card with an notebook PC implements the full functionality based on a set of MPEG-2 video, audio, and system standards. It consists of a dedicated PC-card hardware and a sophisticated PC-software, which are designed and optimized by hardware/software co-design techniques. This PC-card system is good for real-time mobile applications such as video capture and telecommunications.

抄録全体を表示

This paper describes an audio mixing scheme for a software-based H.323 MCU. A load of an audio mixing processing is heavy for a software based MCU. For reducing a load on a software-based MCU, we propose methods of restricting the number of the processing and reducing audio data which is mixed in the processing. We have examined suitable parameters of these methods.

抄録全体を表示

WWW-based high quality video retrieval and transmission system is described. First, a network environment for the video retrieval and transmission service is discussed, and a realization by a 3-layer model is proposed. Based on this model, system and software architecture is developed. Employed security technique is also described.

抄録全体を表示

We have developed a system that creates a TV talk show program from Web-site chat automatically, using the TVML(TV program Making Language)technique. The automatic talk-show generate system has 7characters and 5 sets, and chat users are able to choice a favorate one. First, chat user choose a CG character and start writing with keyboads. After users finish their chat, they can enjoy watching the talk show created by the chat data they made. Many of the users who experiment this system evaluated it very good, so we bilieve that the system is in effect for increasing chat's power of expression.

抄録全体を表示

マルチメディア時代の経済発展を実現するには,著作権等の権利強化指向の法整備と平行して,暗号や電子透かし等のセキュリティ技術を活用した電子的権利処理システムの実現化が重要である.さらに,新たな情報法理念を構築するための本格的研究を開始すべき時期に来ている.NapsterとGnutellaに象徴される情報共有主義動向にも留意する必要があるが,それを禁止すべきか合法とすべきかの2値論理的対処ではなく,法,契約,倫理,技術,組織および教育を含む総合システムによって対処すべきである.

抄録全体を表示

This paper presents a survey of recent technology in the digital watermarking and discusses several problems for attacking the watermark. The robust approaches against Stirmark attack are suggested in the paper.

抄録全体を表示

We can find various rights in Copyright. And we also have neighboring rights which are similar to Copyright. So surveying copyright and neighboring right is useful for us to understand the issue of the Electric Music Distribution. After the surveying, we go ahead in the detail of the "communication to the public" right which is mostly related to the Electric Music Distribution.

抄録全体を表示

In this paper, we will describe copyright and security management in MPEG standard. As well as related standardization such as MPEG-1, MPEG-2, DAVIC, OPIMA, SDMI, MPEG-4/IPMP and MPEG-7, we will report objective and current status of MPEG-21 standardization project.

抄録全体を表示

インターネットや電子商取引の進展に伴い,「暗号」技術の重要性が高まっている.本稿では,この暗号技術の基本機能である「秘匿・認証」技術を概説し,共通鍵暗号・公開鍵暗号それぞれの研究動向について述べる.また,この暗号技術の応用例をいくつか紹介し,さらに,標準化動向についても言及する.

抄録全体を表示