-
Article type: Cover
Pages
Cover1-
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App1-
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Kenji MATSUO, Masayuki HASHIMOTO, Atsushi KOIKE
Article type: Article
Session ID: 1-1
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes an eye extraction method from non-frontal faces based on oriented gradients. The targets of this extraction method are eye corners which can be stably detected from non-frontal faces. Then, the proposed method uses the HOG features to improve the accuracy of eye corner detection. Simulation results confirm the advantages of the proposed method.
View full abstract
-
Takashi KOMATSU, Takahiro SAITO
Article type: Article
Session ID: 1-2
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The total-variation regularization is an effective tool in image processing. Nevertheless, applying the TV regularization to an image contaminated by a large quantity of noise, it will produce a denoised image in which speckled colored-noise is noticeable. This color artifact is caused by the discretization scheme of the TV regularization term, in which simple difference operations between neighboring two pixels are used. We introduce a new multi-resolution scheme into the discretization of the TV regularization term, in which the TV norm is evaluated and minimized at the each resolution image. We present the simulation results and demonstrate superiority of our new approach.
View full abstract
-
Haruya AIZAWA, Daisuke Yamada, Takashi KOMATSU, Takahiro SAITO
Article type: Article
Session ID: 1-3
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper extends the existing TV-G-L^2 nonlinear image decomposition approach to nonlinear color-image decomposition. For utilizing inter-channel color cross-correlations, this paper first introduces the TV norms of color differences and color sums into the TV-G-L^2 functional, and then derives an optimization algorithm using the wavelet transform. Our proposed method decomposes color images without producing color artifacts, and achieves high-quality decomposition robust against random noise.
View full abstract
-
Yuji HORIKAWA, Takashi KOMATSU, Takahiro SAITO
Article type: Article
Session ID: 1-4
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper experimentally compares two mosaicing-and-demosaicing methods with the three-color or the four-color CFA's. As for the three-color CFA, this paper takes up the Alleysson's linear method for the Bayer's CFA, whereas as for the four-color CFA this paper takes up the Hirakawa's method. Computer simulations demonstrate that the former method enhances observation noise. On the other hand, the latter method suppresses observation noise, to some extent.
View full abstract
-
Kensuke HASHIMOTO, Kenji KAMIMURA, Hitomi ITO, Norimichi TSUMURA, Tosh ...
Article type: Article
Session ID: 1-5
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Many methods have been proposed to restore the blurred images caused by motion blur, camera shake, out of focus and so on. However, little work has been done to restore the blurred images by multiple factors. In this paper, we propose a method for restoration of blurred image by camera shake and characteristics of system response.
View full abstract
-
Hiroki SATO, Keiichi KIMURA, Akira KUBOTA, Ryutaro OI, Tadaaki HOSAKA, ...
Article type: Article
Session ID: 1-6
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a method to compose the image of an arbitrary viewpoint from the eight cameras array. The synthesis is based on the reconstruction filter derived from the light field rendering. The image composition can be implemented in real time using an FPGA.
View full abstract
-
Naoki ONO, Kiichi URAHAMA
Article type: Article
Session ID: 1-7
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We formulate the error diffusion processing as the optimization problem and propose an error diffusion halftoning method. The method consists of a cross bilateral filter with an iterative procedure and preserves both smoothness and edge sharpness in an original gray scale image. Experimental results show the proposed method achieves better performance than the standard error diffusion method.
View full abstract
-
Seiji ISHIHARA, Takahiro YAMAMOTO
Article type: Article
Session ID: 1-8
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose an image segmentation method using texture features and Markov random field models in order to extract the contour of a particular segment. Our method consists of coarse region segmentation based on merge and split method and fine modification in the shape of the boundary of a particular segment.
View full abstract
-
Takahiro YAMAMOTO, Seiji ISHIHARA
Article type: Article
Session ID: 1-9
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a new segmentation method for Texture images. Our method extract feature of a texture image by wavelet transformation using wavelet frame, and model the feature based on Markov random field model. By the modeling, we can grasp the feature accurately and concisely. Therefore, we expect to be able to segment the texture image accurately.
View full abstract
-
Yosuke MINE, Takayuki HAMAMOTO
Article type: Article
Session ID: 1-10
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose a new selective visual attention model to extract the region of interest (ROI) for video sequences. In order to extract the ROI, we use multiple salient features such as edge, red-green color contrast, blue-yellow color contrast, and motion, based on the human visual system. By comparative experiments with a human subjective evaluation, we show the effectiveness of the proposed attention model.
View full abstract
-
Masahiro HORIE, Masayuki KASHIMA, Kiminori SATO, Mutsumi WATANABE, Kou ...
Article type: Article
Session ID: 1-11
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose an semi-automatic endocardium tracing method by applying Active Contour Models in two stages. Initial contour have been experimentally prepared for each disease patterns. Experimental result using normal case and three disease cases have shown the effectiveness of the proposed method.
View full abstract
-
Article type: Appendix
Pages
App2-
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Yoshikazu YAJI, Eiji YOKOYAMA, Norihiro WATANABE, Atsushi MICHIMORI, H ...
Article type: Article
Session ID: 2-1
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We examined a response characteristic of a rotating disc under forced vibration. In this paper, we show the following points of rotating disc by forced vibration; a response characteristic of no rotating disc, its vibration mode at a resonant frequency and a response characteristic of a high-speed rotating disc.
View full abstract
-
Chihiro MORITA, Kiyoyasu MARUYAMA, Toshimisu SATO, Naoyuki SAWAMURA, K ...
Article type: Article
Session ID: 2-2
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed collaboration technology between information appliances for user authentication by biometrics via the internet. By this technology, users can enjoy services which require fingerprint authentication via an information appliance without fingerprint sensor, to use an information appliance with the sensor for authentication.
View full abstract
-
Hideki KAWANO, Tetsuro SHIDA, Hideaki KOSAKA
Article type: Article
Session ID: 2-3
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed the PVR (Personal Video Recording) system for TVs with external solid memory device such as SD memory card. Separately from HDD usage for TV as recording device, the small memory device would provide TV users with a slightly different way of conventional recording application.
View full abstract
-
Kensaku SAITOH, Tenryu MISAWA, Keiichi SHIRASUKA, Soichiro MATSUMOTO
Article type: Article
Session ID: 2-4
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In a software verification environment for multimedia equipment, several types of test data like digital multimedia stream are used. Our method is to record these test input data and to replay them in synchronized timing, In this way, test sequences are automatically reproduced.
View full abstract
-
Ryoichi SHIBATA, Haruo OKUDA
Article type: Article
Session ID: 2-5
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have measured transfer rates of three types of PLC systems under the non-recommended conditions such as connecting to UPS output or using simultaneously with different PLC system. Under such conditions, some systems could transfer data.
View full abstract
-
Kensuke UEDA, Satoko MIKI, Takashi SHINOHARA, Eiji MATSUO, Eiji ARITA
Article type: Article
Session ID: 2-6
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We studied on multi-channel recording method for digital broadcasting. Our system configuration and dynamic load sharing method enable 8-channel recording at a lower cost.
View full abstract
-
Kota NAGAI, Tatsuo NOMURA, Masahiko KISHIDA, Naoto HAYASHI
Article type: Article
Session ID: 2-7
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
For the study of a transfer mechanism of signal from magnetic tape to magnetic garnet film, contrast of transferred magnetic domains in garnet will give us meaningful information. We took pictures of magnetic domain patterns by a digital camera and analyzed them in terms of image contrast.
View full abstract
-
Masahiko KISHIDA, Naoto HAYASHI, Katsuhiro IWASAKI, Hiromitsu UMEZAWA, ...
Article type: Article
Session ID: 2-8
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In order to greatly increase the speed at which videotapes are read, we propose an optical reading method using magnetic garnet film. An experiment to reproduce videotape data was carried out at a tape speed of 0.367 m/s, and a CNR of over 30 dB was achieved for a bit-length of 0.385 μm.
View full abstract
-
Hirotaka SHIINO, Mayumi KAWANA, Eiichi MIYASHITA, Sadayuki WATANABE, N ...
Article type: Article
Session ID: 2-9
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Thermally assisted magnetic recording characteristics was investigated for granular perpendicular media. By recording with laser irradiation, output signal increases with laser power and takes maximum value when laser-spot locates under write head.
View full abstract
-
Kentaro Sasaki, Norikazu Kawamura, Haruki Tokumaru
Article type: Article
Session ID: 2-10
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We fabricated a planar lightwave circuit that can align multiple blue-violet beams close together, and constructed optical system with the circuit. Multiple beams from the circuit were evaluated in terms of multibeam light source for optical recording in parallel.
View full abstract
-
Norihiko ISHII, Tetsuhiko MUROI, Nobuhiro KINOSHITA, Koji KAMIJO, Naok ...
Article type: Article
Session ID: 2-11
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Alignments of Focus-lens, recording media, and CCD camera positions are studied. Several rectangle patterns were displayed in SLM and captured by CCD camera. Fourier transform of captured rectangle patterns were calculated and the magnitude of the third order in spatial frequency was picked up with changing the position of the focus-lens. This signal was in good agreement with the signal-to-noise ratio and useful method in holographic data storage.
View full abstract
-
Tetsuhiko MUROI, Nobuhiro KINOSHITA, Norihiko ISHII, Koji KAMIJO, Naok ...
Article type: Article
Session ID: 2-12
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Thick hologram medium, which has possibility of a high data recording density, causes a large amount of shrinkage compared with thin medium under the same shrinkage ratio. Applying adaptive optics with a genetic algorithm to thick medium, intensity distribution in the reconstructed image caused by shrinkage can be improved.
View full abstract
-
Article type: Appendix
Pages
App3-
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Masao ARIZUMI, Naoya SAGARA, Kenji SUGIYAMA
Article type: Article
Session ID: 3-1
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In motion picture coding, inter-picture prediction and cyclic intra-picture coding are used. I-picture spend large amount of bits, and efficiency I-picture is required with keeping its independency. We have proposed a related pre and post-processing which can improve intra-picture coding. Motion compensated inter-frame processing are used in this method. In this report, we apply this method for I-picture. As the results of the experiments using MPEG-4, we recognize the improvement of PSNR significantly.
View full abstract
-
Youhei KASHIMURA, Naoya SAGARA, Kenji SUGIYAMA
Article type: Article
Session ID: 3-2
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In a picture coding, DCT (Discrete Cosine Transform) processing has an advantage in efficiency. However, degradations called block artifact and mosquito noise are appeared in reconstructed pictures. We realize the quantitative estimation method that determines the mosquito blocks. In this method, the ratio of spatial activity is taken between the mosquito block and neighbor flat block. We test the proposed method by using reconstructed pictures which are coded by different quantization scale. We recognize that the results are almost reasonable on subjective picture quality.
View full abstract
-
Kazuo SUGIMOTO, Yoshihisa YAMADA, Yoshiaki KATO
Article type: Article
Session ID: 3-3
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We extended the conventional coset coding to more generic formation to handle more general parameters. The proposed algorithm is applied to intra chroma prediction mode coding in MPEG-4AVC, and the simulation results show significant improvement compared with JM encoder.
View full abstract
-
Yuto KOMATSU, Hiroaki ITO
Article type: Article
Session ID: 3-4
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we describes a new flicker reducing algorithm. The algorithm based on quantization parameter selection amount of motion. The proposed method is reducing flicker power of 21.2% at the maximum.
View full abstract
-
Nao NAKAJIMA, Kazuhisa IGUCHI, Shinichi SAKAIDA, Seiichi GOHSHI
Article type: Article
Session ID: 3-5
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Broadcasters use the 4:2:2 format video recording and editing. It is important to keep the quality of chroma signals. Chroma-key is very common to make programs on TV. We propose AVC/H.264 4:2:2 for contribution. We confirmed availability of applying AVC/H.264 4:2:2 for contribution.
View full abstract
-
Ryo ASATO, Terumasa AOKI, Junji NUMAZAWA
Article type: Article
Session ID: 3-6
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
One of the biggest issues of broadcasting continuous media is how to use bandwidth efficiently and there are many researches about it. However, existing methods can't handle receiving error, so received media must be interrupted by error. In this paper, we propose a new error-resilience scheduling method for efficient broadcasting.
View full abstract
-
Takeharu SHIMIZU, Atsuro ICHIGAYA, Katsunori AOKI, Seiichi GOHSHI, Pet ...
Article type: Article
Session ID: 3-7
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The Super Hi-Vision(SHV) system is expected as the next generation broadcasting service. On the other hand, the IP network which continues to spread is expected to be suitable method for transmitting the SHV signal. So it is important to grasp and analyze the current situation of the international IP network. In this report, we describe the research results of an effective transmission mode and a preliminary transmission experiment in United Kingdom.
View full abstract
-
Toru KIMURA, Satoshi NISHIMURA, Katsunori AOKI
Article type: Article
Session ID: 3-8
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
According to the wide spread of the broadband access environment, many kinds of content distribution services have launched. Broadband IP network is good for VoD service, and, it also provides a potential for expanding service area of broadcast application. The overlay network technology can realize hybrid content delivery network using any IP network. In this report, we consider functional elements and delivery methods of the overlay streaming. Also, we design a prototype of content distribution system and examine its performance by preliminary examination.
View full abstract
-
Satoshi FUJITSU, Hisayuki OHMATA, Kiyohiko ISHIKAWA, Hiroyuki IMAIZUMI
Article type: Article
Session ID: 3-9
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This report presents prototyping of a push-type content delivery system using peer-to-peer networks. In our proposed system the information for controling terminal is distributed anytime by content distributor. Besides it displays states of terminal by an information visualization technique and detect pirates. And content distributor also controls so that they cann't play their contents in their own terminal.
View full abstract
-
Tsuyoshi Yamashita, Masami Ueda, Tetsuo Morita
Article type: Article
Session ID: 3-10
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
For multicast and boradcast contents delivery, it is difficult to retransmit all lost data packets because the packet loss pattern varies from one user to another. The authors have verified that data packet transmission using AL-FEC technology is more efficient than the conventional transmission method by simulations.
View full abstract
-
Toshiharu MORIZUMI, Yumi KIKUCHI, Masahito KAWAMORI, Tomokazu YAMADA, ...
Article type: Article
Session ID: 3-11
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose the seamless contents viewing technique with using IPTV and mobile phone, over the guideline of the Liberty Alliance.
View full abstract
-
Article type: Appendix
Pages
App4-
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Kenichiro ISHIJIMA, Kaibin HOU, Joji Urano, Kiyoyasu ANDO, Takara NAKA ...
Article type: Article
Session ID: 4-1
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed new service "TV Messenger". While watching TV, viewer can talk with their friends on TV sets or One-Seg terminals as well as PCs or mobile-phones.
View full abstract
-
Noriaki FURUSHO
Article type: Article
Session ID: 4-2
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper introduces the quiz form "SAGANMON-Q" on digital terrestrial data broadcasting.
View full abstract
-
Norifumi OKABE, Syunichi OKADA, Atsushi MATSUI
Article type: Article
Session ID: 4-3
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
A video correction system for broadcast materials is described. The system automatically detects human faces in the input video sequence as target areas, then estimates appropriate parameter sets of white/black/gamma levels for the detected targets. The user interface is designed to be simple enough for non-technical users to operate.
View full abstract
-
Ryohei Funatsu, Taku Tsukamoto, Takayuki Imamura, Takayuki Yamashita, ...
Article type: Article
Session ID: 4-4
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed a real-time electronic zoom system to generate an HDTV video from a Super Hi-vision source. The trimming area is selected by a common robot camera remote controller and maximum zoom ratio is 4X. From the results of MTF and S/N measurement, generated HDTV images have minimum required quality for use in HDTV program production.
View full abstract
-
Yoshifumi MATSUSAKI
Article type: Article
Session ID: 4-5
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
For the safe master control operation in digital terrestrial TV broadcasting, the monitoring support system was developed with analyzing Transport Stream.
View full abstract
-
Tomotaka NISHIGAKI, Daiichiro KATO, Hiroyuki OKAMOTO, Jun WAKAKURA
Article type: Article
Session ID: 4-6
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In the latest Digital HD production technique, the development of the camera that can be shot picture images freely in a narrow point was demanded. We at NHK Tokyo have recently produced a new manipulator system with a tiny HD camera.
View full abstract
-
Daiichiro KATO, Hiroyasu MASUDA, Akira KIKKAWA, Kenji MURAKAMI, Masumi ...
Article type: Article
Session ID: 4-7
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed a new type of camera stabilizer which is on shoulder type for broadcasting use. Detecting the vibration of the camera by angular sensors, we control actuators to cancel them. The displacement caused the vibration became almost half, we confirmed that it's effective.
View full abstract
-
Ken-ichi Takahashi, Tsuneo Suzuki
Article type: Article
Session ID: 4-8
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Development of device that records, and accumulates vertical ancillary (VANC) data area used for the information transmission between broadcasting stations.
View full abstract
-
Article type: Appendix
Pages
App5-
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Miwa KATAYAMA, Yuichi IWADATE
Article type: Article
Session ID: 5-1
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We are investigating how to convert 3D models into stereoscopic images of integral photography (IP) using the OpenGL API. This system consists of a high-resolution liquid-crystal panel and a lens array, hi this paper, we discuss the method for converting 3D models into IP images and report experimental results.
View full abstract
-
Yoshiharu MOMONOI, Masahiro SEKINE, Tatsuo SAISHU, Yuzo HIRAYAMA
Article type: Article
Session ID: 5-2
Published: August 01, 2008
Released on J-STAGE: May 24, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed a prototype of a scan-type image capturing system for one-dimensional integral imaging autostereoscopic display. This system is very compact and realizes the fast acquisition of high-density ray. In addition, we established method for evaluation of image quality of this system and evaluated image quality depending on object depth.
View full abstract
