ITU-T T 84-1996 Information Technology - Digital Compression and Coding of Continuous-Tone Still Images Extensions - Terminals for Telematic Services《信息技术 连续色调静止图像的数字压缩和编码 扩展-远程信息处.pdf

《ITU-T T 84-1996 Information Technology - Digital Compression and Coding of Continuous-Tone Still Images Extensions - Terminals for Telematic Services《信息技术 连续色调静止图像的数字压缩和编码 扩展-远程信息处.pdf》由会员分享，可在线阅读，更多相关《ITU-T T 84-1996 Information Technology - Digital Compression and Coding of Continuous-Tone Still Images Extensions - Terminals for Telematic Services《信息技术 连续色调静止图像的数字压缩和编码 扩展-远程信息处.pdf（85页珍藏版）》请在麦多课文档分享上搜索。

1、INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU TERMINALS FOR TELEMATIC SERVICES T.84 (07/9 6) INFORMATION TECHNOLOGY - OF CONTINUOUS-TONE STILL IMAGES: DIGITAL COMPRESSION AND CODING EXTENSIONS ITU-T Recommendation T.84 (Previously “CCITT Recommendation”)

2、 - STD*ITU-T RECMN T-8q-ENGL 199b YBb259L Ob29145 951 FOREWORD ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (IT-T) is a permanent organ of the IT. Some 179 member countries, 8

3、4 telecom operating entities, 145 scientific and industrial organizations and 38 international organizations participate in ITU-T which is the body which sets world telecommunications standards (Recommendations). The approval of Recommendations by the Members of ITU-T is covered by the procedure lai

4、d down in WTSC Resolution No. 1 (Helsinki, 1993). In addition, the World Telecommunication Standardization Conference (WTSC), which meets every four years, approves Recommendations submitted to it and establishes the study programme for the following period. In some areas of information technology w

5、hich fall within ITU-Ts purview, the necessary standards are prepared on a collaborative basis with IS0 and IEC. The text of IT-T Recommendation T.84 was approved on 3rd of July 1996. The identical text is also published as ISO/IEC International Standard 10918-3. NOTE In this Recommendation, the exp

6、ression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. O IT 1997 All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopyi

7、ng and microfilm, without permission in writing from the ITU, except as noted in footnote 1) in Annex F. CONTENTS Page 1 Scope 2 Normative references . 3 Definitions, abbreviations, symbols and conventions 4 General . 5 Compressed data format requirements . 6 Encoder requirements . 7 Decoder require

8、ments Annex Annex Annex Annex Annex Annex Annex Annex Annex A . Mathematical definitions B - Compressed data formats . C - Variable quantization . D - Selective refinement . E - Tiling F - Still Picture Interchange File Format (SPIFF) G - Compliance testing . H - Examples and guidelines I - Bibliogr

9、aphy ITU-T Rec . T.84 (1996 E) 1 1 2 6 10 10 11 12 14 28 31 34 39 57 72 80 1 STD - ITU-T RECMN T.84-ENGL 724 9 Summary This Recommendation I International Standard, Digital compression and coding of continuous-tone still images, is published as three parts: - - - CCITT Rec. T.81 i ISO/IEC 10918-1: R

10、equirements and guidelines; IW-T Rec. T.83 I ISOLEC 10918-2: Compliance testing; IW-T Rec. T.84 I ISOAEC 10918-3: Extensions. This Recommendation I International Standard sets out requirements and guidelines for encoding and decoding extensions to the processes defined by CCITT Rec. T.81 I ISO/IEC 1

11、0918-1, and for the coded representation of compressed image data of these extensions. This Recommendation I International Standard also defines tests for determining whether implementations comply with the requirements for the various encoding and decoding extensions. This Recommendation I Internat

12、ional Standard: - defines extensions including variable quantization, selective refinement, composite tiling, and a Still Picture Interchange File Format (SPIFF) to processes for converting source image data to compressed image data; defines extensions to processes for converting compressed image da

13、ta to reconstructed image data; defines coded representations for compressed image data; gives guidance and examples on how to implement these extensions in practice; describes compliance tests for these extensions. - - - - Patents ISOAEC JTCl/SC29/WGI and ITU-T TSB are not aware of any patent restr

14、ictions applying to the extensions in this Recommendation I International Standard. 11 ITU-T Rec. T.84 (1996 E) INTERNATIONAL STANDARD ITU-T RECOMMENDATION INFORMATION TECHNOLOGY - DIGITAL COMPRESSION AND CODING OF CONTINUOUS-TONE STILL IMAGES: EXTENSIONS 1 Scope This Recommendation I International

15、Standard is applicable to continuous-tone - grayscale or colour - digital stili image data. It is applicable to a wide range of applications which require use of compressed images. This Recommendation I International Standard: - defines extensions including variable quantization, selective refinemen

16、t, tiling, and a Still Picture Interchange File Format (SPIFF) to processes for converting source image data to compressed image data; defines extensions to processes for converting compressed image data to reconstructed image data; defines coded representations for compressed image data; gives guid

17、ance and examples on how to implement these extensions in practice; describes compliance tests for these extensions. - - - - 2 Normative references The following Recommendations and International Standards contain provisions which, through references in this text, constitute provisions of this Recom

18、mendation I International Standard. At the time of publication, the editions indicated were valid. AI1 Recommendations and Standards are subject to revision, and parties to agreements based on this Recommendation I International Standard are encouraged to investigate the possibility of applying the

19、most recent edition of the Recommendations and Standards listed below. Members of IEC and IS0 maintain registers of currently valid International Standards. The Telecommunication Standardization Bureau (TSB) of the IT maintains a list of currently valid IT-T Recommendations. 2.1 Identical Recommenda

20、tions I International Standards - ITU-T Recommendation H.262 (1995) I ISOAEC 138 18-2995, Information technology - Generic coding of moving pictures and associated audio information: Video. - CCITT Recommendation T.81 (1992) I ISO/IEC 109 18-1 : 1994, Information technology - Digital compression and

21、 coding of continuous-tone still images - Requirements and guidelines. - IT-T Recommendation T.82 (1993) I ISO/LEC 11544:1993, Information technology - Coded representation of picture and audio information - Progressive bi-level image compression. ITU-T Recommendation T.83 (1994) I ISO/IEC 10918-2:1

22、995, Information technology - Digital compression and coding of continuous-tone still images: Compliance testing. - 2.2 Additional references - - IS0 3166: 1993), Codes for the representation of names of countries. IS0 5807: 1985, Information processing - Documentation symbols and conventions for da

23、ta, program and systemjlowcharts, program nemork charts and system resources charts. IS0 8601:1988, Data elements and interchange formats - Information interchange - Representation of dates and times. - I) Currently under revision. ITU-T Rec. T.84 (1996 E) 1 - IS0 8859-1: 1987, Information processin

24、g - no new processes are defined. This Recommendation I International Standard also defines extensions to the compressed data formats, i.e. interchange format and the abbreviated formats. Each encoding or decoding extension shall only be used in combination with particular coding processes and only

25、in accordance with the requirements set forth herein. These extensions are backward compatible in the sense that decoders which implement these extensions will also support configuration subsets that are currently defined by CCITT Rec. T.81 I ISO/IEC 10918-1. 4.1 The following extensions are specifi

26、ed: Extensions specified by this Recommendation I International Standard - An extension which provides for variable quantization within a scan. The variable quantization extension may be used in conjunction with any of the DCT-based processes with the exception of the baseline process. The variable

27、quantization extension provides for scaling of all quantization tables at the 8 x 8 block level; An extension which provides for selective refinement. The selective refinement extension refers to selecting rectangular regions of an image components for further refinement. There are several types of

28、selective refinement: a) The first type of selective refinement, referred to as hierarchical selective refinement, allows only a region of one or more components to be further refined by the next differential frame of a hierarchical sequence. - 6 ITU-T Rec. T.84 (1996 E) STD- ITU-T RECMN T.84-ENGL L

29、79b 4862593 Ob24354 Yb4 M ISO/IEC10918-3: 1997(E) b) The second type of selective refinement, referred to as progressive selective refinement, applies to the DCT-based progressive mode of operation. This type of Selective refinement allows more non- zero DCT coefficients, more bits to the DCT coeffi

30、cients, or both, to be added to a region of one or more component. The third type of selective refinement, referred to as component selective refinement, is used to specify a region of an image, which contains colour components which do not exist in other regions of the image. c) - An extension whic

31、h provides for tiling. The tiling extension is used to associate a number of sub-images, also called tiles, in order to form a single tiled image. The three types of tiling are summarized below: a) For simple tiling, all tiles except possibly those on the right and bottom border have the same maximu

32、m dimensions, number of components, component IDS and scaling factors. Tiles (i.e. their components arrays) are non-overlapping and contiguous. The tiles are coded sequentially from left-to-right and top-to-bottom. For pyramidal tiling, multiple resolution versions of the same image (i.e. each versi

33、on has different maximum dimensions) may be stored together in the same data stream. Each version of the image (also called a “resolution levei”) is stored as a tiled image with simple tiling. For composite tiling, there are no restrictions except that all tiles shall have the same component identif

34、iers. b) c) - An extension which provides for the interchange of compressed image files between application environments. This extension is referred to as the Still Picture Interchange File Format (SPIFF) extension. Other extensions include addition of a version number to the compressed data format

35、and increasing the limit on the number of data units in a minimum coded unit to 20. - The following subclauses describe these extensions in greater detail. 4.1.1 Variable quantization extension The variable quantization extension is an enhancement to the quantization procedure of DCT-based processes

36、 which provides for changes to the quantization table values within a scan at the 8 x 8 block level. This extension may be used in conjunction with any of the DCT-based processes with the exception of the Baseline Process. The quantization procedure as defined by CCITT Rec. T.81 I ISOEC 10918-1 is t

37、he step in the encoding process where each of the 64 DCT coefficients are quantized using one of 64 corresponding values from a quantization table. CCITT Rec. T.81 I ISOIEC 10918-1 permits quantization tables to be redefined prior to the start of a scan but does not allow quantization table values t

38、o be changed within a scan. The variable quantization extension defined by this Recommendation I International Standard provides for scaling of quantization values at the 8 x 8 block level. The variable quantization extension introduces a quantizer scale factor which may be coded in the compressed d

39、ata stream at the start of any 8 x 8 block. The quantizer scale factor is used to scale the quantization table values which correspond to the AC coefficients in the quantization procedure. All defined quantization tables are scaled by the same quantizer scale factor. This extension provides the foll

40、owing capabilities: - The ability to compress an image to less than a bounded size with a single sequential pass over the image. The capability is valuable to applications which utilize a fixed-size compressed picture memory. The ability to use the masking properties of the human visual system more

41、effectively, and thereby achieve greater compression rates for the same subjective quality. The ability to transcode, i.e. entropy decoding followed by entropy encoding, between some coded data representations defined by this Recommendation I International Standard and those defined by some other st

42、andards, including ISOAEC 11 172-2 and ITU-T Rec. H.262 I ISO/IEC 13818-2 (WEG). - - 4.1.2 Selective refinement extension The selective refinement extension is used to select a rectangular region of one or more components of an image for further refinement. The different types of selective refinemen

43、t are described below. ITU-T Rec. T.84 (1996 E) 7 STD-ITU-T RECMN T-BY-ENGL L77b m q8b2573 Ob2LiL55 BTU ISOAEC 10918-3 : 1997 (E) 4.1.2.1 Hierarchical selective refinement Hierarchical selective refinement is used in the hierarchical mode of operation for refining a rectangular region of one or more

44、 components of an image. The location of the region of each component of the image to be selectively refined is specified immediately prior to a differential frame within an hierarchical sequence. The size of the region of each component is specified in the differential frame header. The difference

45、image data reconstructed from the differential frame is then added only to the specified region of each component. One of the primary uses for this type of selective refinement is for coding a particular region of interest with greater detail than the remainder of the image. 4.1.2.2 Progressive sele

46、ctive refinement The second type of selective refinement, referred to as progressive selective refinement, is used in the DCT-based progressive mode of operation. This type of selective refinement is used for similar reasons as hierarchical selective refinement. Progressive selective refinement may

47、be applied to the DCT-based processes which use spectral selection, successive approximation, or both procedures in combination. When progressive selective refinement is applied to a scan which uses the spectral selection procedure, more non-zeroDCT coefficients are added to a region of each compone

48、nt. When it is applied to a scan which uses the successive approximation procedure, more bits are added to the DCT coefficients in a region of each component. Progressive selective refinement may also be applied to scans which use both procedures. In ail cases, the location and size of the region of

49、 each component to be selectively refined is specified immediately prior to the scan which is used for selective refinement. 4.1.2.3 Component selective refinement The third type of selective refinement, referred to as component selective refinement, may be used in all modes of operation for specifying a region of an image which contains colour components that do not exist in other regions of the image. The most common use for this type of selective refinement is for representing images which are mixed grayscale and colour. 4.1.3 Tiling extension The tiling extension is u