All issues
Predecessor
26.76
Displaying 1-12 of 12 articles from this issue
- |<
- <
- 1
- >
- >|
-
Article type: Cover
Pages Cover1-
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (15K) -
Article type: Index
Pages Toc1-
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (45K) -
Kunitoshi KOMATSU, Kaoru SEZAKIArticle type: Article
Session ID: BCS2002-47/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSTwo-point lossless rotation transforms are used on realizing lossless DCTs or lossless orthogonal wavelet transforms. However, two-point lossless rotation transforms based on the ladder network do not have high compatibility with their non-lossless versions. Therefore, in this paper, we propose two-point lossless rotation transforms with periodic structure and compare them with those based on the ladder network. We design the corresponding tables by making use of periodic structure after supplementary rotation transforms or supplementary widening conversions without replacing the transform matrix with the integer transform matrix. Accordingly, the proposed method is simple and the size of corresponding table is smaller than that of the conventional method. Since we can design the corresponding tables in consideration of compatibility with the non-lossless transforms, we can design the lossless rotation transforms that have high compatibility with them. Computer simulations show that the proposed methods have higher compatibility with non-lossless rotation transforms than the methods based on the ladder network.View full abstractDownload PDF (723K) -
Ken NAKAMURA, Takeshi YOSHITOME, Yoshiyuki YASHIMAArticle type: Article
Session ID: BCS2002-48/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWhen Super High Resolution image and stereo image which consists of multiple video channels is encoded and transmitted by multiple MPEG codecs, it is very important to synchronize the multiple video channels. We have developed a video frame synchronizer for multi-channel video codec transmission system which utilizes existing MPEG-2 HDTV codecs. This system imposes markers on the top or bottom lines of images for all video channel before encoding and read the markers from the decoded images and automatically synchronize all channels. We constructed a multi-channel video codec transmission system using this synchronizer and transmited stereo HDTV image and Super High Resolution image over ATM and verified the performance of this system.View full abstractDownload PDF (909K) -
Yasuo SAMBE, Shintaro WATANABE, Dong YU, Taichi NAKAMURA, Naoki WAKAMI ...Article type: Article
Session ID: BCS2002-49/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSA high speed video transcoding sytem with distributed computing is described. The system can simultaneously transcode a high quality MPEG-2 coded video source to several kinds of video that may differ in conding method, bit-rate. We focused on how to divide and transcode the video source suitably for distributed transcoding, and reallocate bits in order to enhance the quality of transcoded video.View full abstractDownload PDF (833K) -
Noriyuki SHIRAI, Ichiro MATSUDA, Susumu ITOHArticle type: Article
Session ID: BCS2002-50/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe previously proposed a novel lossless image coding scheme using minimum-rate predictors and variable length codes (VLCs). In this scheme, linear predictors are designed for each image so that a total coding rate estimated on Gaussian models can have a minimum. Therefore coding performance deteriorates when probability density functions (PDFs) of actual prediction errors do not conform to the models. To cope with this problem, in this paper, we introduce new PDF models based on generalized Gaussian functions and iteratively optimize not only predictors but also parameters of the models. Moreover, adaptive arithmetic code is employed in stead of the VLCs to improve coding efficiency in flat areas of images. Simulation results indicate that the proposed scheme achieves higher coding performance than the state-of-the-art lossless coding schemes for most natural images.View full abstractDownload PDF (849K) -
Seiichi UCHIDA, Hiroaki SAKOEArticle type: Article
Session ID: BCS2002-51/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSFor efficient video compression, a pixel-based motion compensation technique where motion is determined at every pixel is investigated. The pixel-based motion compensation technique is expected to provide better performance on the minimization of prediction error than conventional block-based motion compensation techniques. In order to suppress the increase of motion information, several constraints are imposed on the motions of neighboring pixels. Experimental results have shown that the present technique can attain higher compression rates than a block-based technique for image sequences with large motion.View full abstractDownload PDF (1030K) -
Yuji ITO, Mutsumi WATANABEArticle type: Article
Session ID: BCS2002-52/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThe automatic stereo calibration which computes the fundamental matrix showing the relative position information between camera was newly developed. In the system, the feature points, from one pair of stereo pictures, are extracted using the Harris Corner Detector and matched for creating 3D environment model. Feature points used for stereo calibration were automatically selected based on the reliability estimation using brightness and feature information around each feature points. Furthermore, for the purpose of improvement in accuracy of stereo calibration, the fundamental matrix was calculated using the corresponding points in the position distant on the picture. This system was installed on PC as a program using VC++. Experiment result using indoor scenes showed the effectiveness of the system.View full abstractDownload PDF (969K) -
Yu WANG, Kazuhiro TAKEDA, Kiminori SATO, Shigeru NAKAYAMAArticle type: Article
Session ID: BCS2002-53/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe have examined the human recognition method using eigenear images. We have found that the recognition method by eigenface images with principal component analysis is effective in the eigenear images. The ear images in the experiment come from 110 persons' data collected under the fixed condition. They have 6 different images for each person. We used 5 of 6 images for learning, and the last one is used for test pattern. In this experiment, we have examined the relationships between several numbers of the pattern, dimensional numbers of the eigenear, size of the mosaic, and recognition rate. In conclusion we got the high recognition rate. We found that the normalization of the pixels in the image results in the improvement of the recognition rate.View full abstractDownload PDF (722K) -
Hiroyuki MARUO, Mutsumi WATANABEArticle type: Article
Session ID: BCS2002-54/AIT2002-1
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSAn automatic region segmentation method based on the mixture Gaussian model was newly developed. First, distribution parameters that correspond to objects in the scene are estimated. The parameters are estimated from hue histogram data using the EM algorithm. Next, range parameters for region segmentation are determined using the estimated distribution parameters. Finally, the images are segmented by these range parameters. Experimental results of indoor and out door scenes usiner the PC oroeram showed the effectiveness of the method.View full abstractDownload PDF (913K) -
Article type: Appendix
Pages App1-
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (74K) -
Article type: Appendix
Pages App2-
Published: November 21, 2002
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (74K)
- |<
- <
- 1
- >
- >|