1、INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STAN DARD IZATI ON SECTOR OF ITU T.88 (02/2 o 00) SERIES T: TERMINALS FOR TELEMATIC SERVICES Information technology - Lossy/lossless coding of billeve1 images ITU-T Recommendation T.88 (Previously CCIlT Recommendation) I“l?ERNATIONAL STAN
2、DARD ISO/CEI 14492 ITU-T RECOMMENDATION T.88 INFORMATION TECHNOLOGY - LOSSYlLOSSLESS CODING OF BI-LEVEL IMAGES Summary This Recommendation I International Standard, commoniy known as JBIG2, defines a coding method for bi-level images (e.g. black and white printed matter). These are images consisting
3、 of a single rectangular bit plane, with each pixel taking on one of just two possible colours. This Recommendation I Internatil Sandard has been explicitly prepared for a lossy, lossless, and lossy-to-lossless image compression. Source ITU-T Recommendation T.88 was approved on 10 February 2000. An
4、identical text is also published as ISO/IEC 14492. ITU-T Rec. T.88 (2000 E) 1 FOREWORD ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommuni- cations. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU.
5、The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topic
6、s for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1. In some areas of information technology which fall within ITU-Ts purview, t
7、he necessary standards are prepared on a collaborative basis with IS0 and IEC. NOTE In this Recommendation, the expression “Administration“ is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. INTELLECTUAL PROPERTY RIGHTS The ITU draws attent
8、ion to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or oth
9、ers outside of the Recommendation development process. As of the date of approval of this Recommendation, the ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not repr
10、esent the latest information and are therefore strongly urged to consult the TSB patent database. o ITU 2002 Ail rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission
11、in writing fiom the ITU. 11 ITU-T Rec. T.88 (2000 E) CONTENTS O Introduction Interpretation and use of the requirements 0.1.1 Subject matter for JBIG2 coding . 0.1.2 Relationship between segments and documents 0.1.3 Structure and use of segments . 0.1.4 Internal representations . O . 1.5 Decoding re
12、sults O . 1.6 Decoding procedures . 0.2 Lossy coding . 0.2.1 Symbol coding . 0.2.2 Generic coding 0.2.3 Halftone coding . 0.2.4 Consequences of inadequate segmentation . 1 Scope 2 Normative References 3 Terms and Dennitions 4 Symbols and Abbreviations 4.1 Abbreviations 4.2 Symbol dennitions 4.3 Oper
13、ator dennitions . 5 Conventions 5.1 Typographic conventions 5.2 Binary notation 5.3 Hexadecimal notation 5.4 Integer value syntax 5.4.1 Bit packing 5.4.2 Multi-byte values . 5.4.3 Bit numbering 5.4.4 Signedness . Amy notation and conventions 6 Decoding Procedures Introduction to decoding procedures
14、. Generic region decoding procedure . 6.2.1 General description 6.2.2 Input parameters 6.2.3 Return value 6.2.4 Variables used in decoding 6.2.5 Decoding using a template and arithmetic coding . 6.2.6 Decoding using MMR coding . 0.1 5.5 5.6 Image and bimap conventions 6.1 6.2 6.3 Generic Refinement
15、Region Decoding Procedure . 6.3.1 General description 6.3.2 Input parameters 6.3.3 Return value 6.3.4 Variables used in decoding 6.3.5 Decoding using a template and arithmetic coding . Page vii vii vii . M11 M11 M11 X X . . xi xi xi xii xii 1 1 1 3 3 4 10 10 10 10 11 11 11 11 11 11 11 11 12 12 13 13
16、 13 13 14 14 18 19 19 19 19 20 20 . ITU-T Rec . T.88 (2000 E) 111 6.4 6.5 6.6 6.7 Text Region Decoding Procedure . 6.4.1 General description 6.4.2 Input parameters 6.4.3 Return value 6.4.4 Variables used in decoding 6.4.5 Decoding the text region . 6.4.6 Strip delta T . 6.4.7 First symbol instance S
17、 coordinate 6.4.8 6.4.9 Symbol instance T coordinate . 6.4.1 O 6.4.1 1 Symbol instance bitmap Symbol Dictionary Decoding Procedure . 6.5.1 General description 6.5.2 Input parameters 6.5.3 Return value 6.5.4 6.5.5 6.5.6 Height class delta height 6.5.7 Delta width 6.5.9 6.5.1 O Exported symbols Halfto
18、ne Region Decoding Procedure . 6.6.1 General description 6.6.2 Input parasneters 6.6.3 Return value 6.6.4 Variables used in decoding 6.6.5 Pattern Dictionary Decoding Procedure 6.7.1 General description 6.7.2 Input parameters 6.7.3 Return value 6.7.4 Variables used in decoding 6.7.5 Decoding the pat
19、tern dictionary Subsequent symbol instance S coordinate . Symbol instance symbol ID . Variables used in decoding Decoding the symbol dictionary 6.5.8 Symbol bitmap Height class collective bitmap . Decoding the halftone region 7 Control Decoding Procedure 7.1 7.2 7.3 7.4 iv General description . Segm
20、ent header syntax 7.2.1 Segment header fields 7.2.2 Segment numb er 7.2.4 7.2.5 7.2.6 Segment page association 7.2.7 Segment data length Segment types . 7.3.1 Rules for segment references . 7.3.2 Segment syntaxes 7.4.1 Region segment information field . 7.4.2 Symbol dictionary segment syn. 7.4.3 7.4
21、.4 Pattern dictionary segment syntax . 7.4.5 7.4.6 Generic region segment syntax 7.4.7 Generic refinement region syntax 7.4.8 Page information segment syntax 7.2.3 Segment header flags . Referred-to segment count and retention flags Referred-to segment numb eTs 7.2.8 Segment header example . Rules f
22、or page associations Text region segment syntax . Halftone region segment syntax ITU-T Rc . T.88 (2000 E) Page 23 23 23 24 24 25 28 28 28 29 29 29 30 30 30 30 30 32 34 34 34 37 37 38 38 38 39 39 39 42 42 42 42 43 43 44 44 45 45 45 45 45 47 47 47 47 48 49 50 50 50 51 56 66 67 70 72 73 Page 7.4.9 End
23、of page segment syntax 76 7.4.10 End of stripe segment syntax . 76 7.4.11 End of file segment syntax 76 7.4.12 Profiles segment syntax . 76 7.4.13 Code table segment syntax 77 7.4.14 Extension segment syntax . 77 7.4.15 Defined extension types 77 8 Page Make-up . 78 8.1 Decoder model 78 8.2 Page ima
24、ge composition 78 82 A . 1 General description . 82 A.2 Procedure for decoding values (except IAID) . 82 A.3 The MID decoding procedure . 84 Annex A - Arithmetic Integer Decoding Procedure . Annex B . Hu- Table Decoding Procedure B.l General description . B.2 Code table structure . B.2.1 Code table
25、flags . B.2.2 Code table lowest value . B.2.3 Code table highest value B.3 Assigning the prefix codes B.4 Using a HufEnan table B.5 Standard Huffman tables . 86 86 86 87 87 87 87 88 89 Annex C . Gray-scale image Decoding Proceure 97 C.l General description . 97 C.2 Input parameters 97 C.3 Return val
26、ue 97 C.4 Variables used in decoding . 97 C.5 Decoding the gray-scale image . 98 Annex D . File Formats D.l Sequential organisation . D.2 Random-access organisation . D.3 Embedded organisation . D.4 File header syntax . D.4.1 ID string . D.4.2 File header flags D.4.3 Number ofpages . Annex E - Arith
27、metic Coding . E.l Binary encoding E.l.l Recursive interval subdivision . E.1.2 Coding conventions and approximati0 ns . Description ofthe arithmetic encoder . E.2.1 Encoder code register conventions E.2.2 Encoding a decision (ENCODE) . E.2.3 Encoding a 1 or O (CODE1 and CODEO) E.2.4 Encoding an MPS
28、 or LPS (CODEMPS and CODELPS) E.2.5 Probability estimation E.2.6 Renormalisation in the encoder (RENORME) E.2.7 Compressed data output (BYTEOUT) E.2.8 Initialisation of the encoder (INITENC) E.2.9 E.2.10 Minimisation ofthe compressed data . E.2 Termination of encoding (FLUSH) . 99 99 99 100 100 100
29、100 100 101 101 101 101 102 103 103 103 104 105 105 106 107 107 107 V ITU-T Rc . T.88 (2000 E) E.3 Arithmetic decoding procedure . Decoder code register conventions Decoding a decision (DECODE) . Renormalisation in the decoder (RENORMD) Compressed data input (BYTEIN) Initialisation of the decoder (I
30、NITDEC) Resynchronisation of the decoder Resetting arithmetic coding statistics Saving arithmetic coding statistics Annex F - Profiles . Annex G -Arithmetic Decoding Procedure (Software Conventions) . Annex H - Datastream Example and Test Sequence . Datastream example Test sequence for arithmetic co
31、der Bibliography E.3.1 E.3.2 E.3.3 E.3.4 E.3.5 E.3.6 E.3.7 E.3.8 H . 1 H.2 Page 109 111 111 111 111 114 114 115 115 116 119 121 121 142 147 vi ITU-T Rc . T.88 (2000 E) O Introduction This Recommendation I International Standard, informally called JBIG2, defines a coding method for bi-level images (e
32、.g. black and white printed matter). These are images consisting of a single rectangular bit plane, with each pixel taking on one ofjust two possible colours. Multiple colours are to be handled using an appropriate higher level standard such as ITU-T Recommendation T.44. It is being drafted by the J
33、oint Bi-level Image Experts Group (JBIG), a “Collaborative Team“, established in 1988, that reports both to ISOIEC JTC I/SC29/WGI and to ITU-T. Compression of this type of image is also addressed by existing facsimile standards, for example by the compression algorithms in ITU-T Recommendations T.4
34、(MH, MR), T.6 (MMR), T.82 (JBIGl), and T.85 (Application profile of JBIG1 for facsimile). Besides the obvious facsimile application, JBIG2 will be useful for document storage and archiving, coding images on the World Wide Web, wireless data transmission, print spooling, and even teleconferencing. As
35、 the result of a process that ended in 1993, JBIG produced a first coding standard formally designated IT-T Recommendation T.82 I International Standard ISOIEC 11544, which is informally known as JBIG or JBIG1. JBIG1 is intended to behave as lossless and progressive (lossy-to-lossless) coding. Thoug
36、h it has the capability of lossy coding, the lossy images produced by JBIGI have significantly lower quality than the original images because the number of pixels in the lossy image cannot exceed one quarter of those in the original image. On the contrary, BIG2 was explicitly prepared for lossy, los
37、sless, and lossy-to-lossless image compression. The design goal for JBIG2 was to allow for lossless compression performance better than that of the existing standards, and to allow for lossy compression at much higher compression ratios than the lossless ratios of the existing standards, with almost
38、 no visible degradation of quality. In addition, BIG2 allows both quality-progressive coding, with the progression going fiom lower to higher (or lossless) quality, and content-progressive coding, successively adding different types of image data (for example, first text, then halftones). A typical
39、JBIG2 encoder decomposes the input bi-level image into several regions and codes each of the regions separately using a different coding method. Such content-based decomposition is very desirable especially in interactive multimedia applications. JBIG2 can also handle a set of images (multiple page
40、document) in an explicit manner. As is typical with image compression standards, JBIG2 explicitly defines the requirements of a compliant bitstream, and thus defines decoder behaviour. JBIG2 does not explicitly define a standard encoder, but instead is flexible enough to allow sophisticated encoder
41、design. In fact, encoder design will be a major differentiator among competing BIG2 implementations. Although this Recommendation I International Standard is phrased in terms of actions to be taken by decoders to interpret a bitstream, any decoder that produces the correct result (as defined by thos
42、e actions) is compliant, regardless of the actions it actually takes. Annexes A, B, C, D, E, and F are normative, and thus form an integral part of this Recommendation I Intemational Standard. Annexes G and H are informative, and thus do not form an integral part of this Recommendation 1 Internation
43、al Standard. 0.1 This section is informative and designed to aid in interpreting the requirements of this Recommendation I Intemational Standard. The requirements are written to be as general as possible to allow a large amount of implementation flexibility. Hence the language of the requirements is
44、 not specific about applications or implementations. In this section a correspondence is drawn between the general wording of the requirements and the intended use of this Recom- mendation I Intemational Standard in typical applications. Interpretation and use of the requirements 0.1.1 JBIG2 is used
45、 to code bi-level documents. A bi-level document contains one or more pages. A typical page contains some text data, that is, some characters of a small size arranged in horizontal or vertical rows. The characters in the text part of a page are called symbols in JBIG2. A page may also contain “halft
46、one data“, that is, gray-scale or colour multi- level images (e.g. photographs) that have been dithered to produce bi-level images. The periodic bitmap cells in the halftone part of the page are called patterns in JBIG2. In addition, a page may contain other data, such as line art and noise. Such no
47、n-text, non-halftone data is called generic data in JBIG2. The JBIG2 image model treats text data and halftone data as special cases. It is expected that a BIG2 encoder will divide the content of a page into a text region containing digitised text, a halltone region containing digitised halftones, a
48、nd a generic region containing the remaining digitised image data, such as line-art. In some circumstances, it is better (in image quality or compressed data size) to consider text or halftones as generic data; conversely, in some circumstances it is better to consider generic data using one of the
49、special cases. Subject matter for JBIG2 coding ITU-T RW. T.88 (2000 E) vii An encoder is permitted to divide a single page into any number of regions, but often three regions will be sufficient, one for textual symbols, one for halftone patterns, and the third for the generic remainder. In some cases, not all types of data may be present, and the page may consist of fewer than three regions. The various regions may overlap on the physical page. BIG2 provides the means to specify how the overlapping regions are recombined to form the final page image. A text region consists of a num
