Project/Area Number |
60850171
|
Research Category |
Grant-in-Aid for Developmental Scientific Research
|
Allocation Type | Single-year Grants |
Research Field |
電子通信系統工学
|
Research Institution | KOGAKUIN UNIVERSITY |
Principal Investigator |
MINAMI Toshi Prof. of KOGAKUIN Univ., 工学部, 教授 (80146729)
|
Co-Investigator(Kenkyū-buntansha) |
KURUMA Koji Directer, Image Comminication Development Divison OKI Electric Industry Co.,Ltd., 画像通信技術部, 部長
YAMAZAKI Yasuhiro Manager, Terminal Equipment Lab. KDD, 研究所端末装置研究室, 室長
SHINOHARA Katsuyuki Assistant of KOGAKUIN Univ., 工学部, 助手 (40100309)
NAKAMURA Osamu Lecturer of KOGAKUIN Univ., 工学部, 講師 (70100336)
TAKAHASHI Shizuaki Assistant Prof. of KOGAKUIN Univ., 工学部, 助教授 (90100304)
|
Project Period (FY) |
1985 – 1986
|
Project Status |
Completed (Fiscal Year 1986)
|
Budget Amount *help |
¥5,000,000 (Direct Cost: ¥5,000,000)
Fiscal Year 1986: ¥5,000,000 (Direct Cost: ¥5,000,000)
|
Keywords | Mixed mode communication / Block truncation coding / Vector quantization / Color image / Image processing / ミクストモード / 可変ブロックサイズ符号化 / 階調画像 / コードワード / 部分サンプリング / 平面近似 |
Research Abstract |
The mixed mode communication system which integrates the teletex and the facsimile is considered to be the most useful transmission system from the viewpoint of the transmission efficiency and the quality of received documents. To implement the mixed mode communication system, researches and developments are being carried out extensively in various laboratories. In this system the textual part of a document is transmitted with data codes and the graphic part is transmitted with facsimile codes and the amount of the coded signal can be compared effectively. On the other hand, in proportion to the increase of visual communication demands, the requirements for functions of facsimile are diversified and complicated. Among them, the development of encoding system for the photograph facsimile and color facsimile becomes very important subject. Under this background, in this study, the high compression encoding system with (1) variable size block coding scheme based on block merging and (2) successive clustering and predictive vector quantization, is studied theoretically and experimentally. An experimental model which consists of an input camera, a super personal computer and a display terminal is fabricated and the technical feasibility and usefulness of the coding algorithms are investigated. The experimental results of encoding show that for the continuous tone images, the number of required bits for encoding are 0.43-0.66bits/pel and the S/N of the decoded images are 38.29-34.18dB with the algorithm (1), and with the algorithm (2) those values are 0.5-0.8bits/pel and 30-36dB. This shows that the coding efficiency is increased about 30% compared with the conventional coder. For the colored images, the number of required bits for encoding and the S/N are 0.6-1.02bits/pel and 38.5-33.3dB with the algorithm (1) and 0.3-3.30bits/pel and 34-48.4dB with the algorithm (2). These results categorically prove the usefulness and superiority of the newly developed encoding system.
|