ISO/IEC 10918-1 : 1993(E)
Annex J
Hierarchical mode of operation
(This annex forms an integral part of this Recommendation | International Standard)
This annex provides a functional specification of the coding processes for the hierarchical mode of operation.
In the hierarchical mode of operation each component is encoded or decoded in a non-differential frame. Such frames may
be followed by a sequence of differential frames. A non-differential frame shall be encoded or decoded using the
procedures defined in Annexes F, G and H. Differential frame procedures are defined in this annex.
The coding process for a hierarchical encoding containing DCT-based processes is defined as the highest numbered
process listed in Table J.1 which is used to code any non-differential DCT-based or differential DCT-based frame in the
compressed image data format. The coding process for a hierarchical encoding containing only lossless processes is
defined to be the process used for the non-differential frames.
Table J.1 Coding processes for hierarchical mode
Process
Non-differential frame specification
1
1
Extended sequential DCT, Huffman, 8-bit
Annex F, process 2
1
2
Extended sequential DCT, arithmetic, 8-bit
Annex F, process 3
1
3
Extended sequential DCT, Huffman, 12-bit
Annex F, process 4
1
4
Extended sequential DCT, arithmetic, 12-bit
Annex F, process 5
1
5
Spectral selection only, Huffman, 8-bit
Annex G, process 1
1
6
Spectral selection only, arithmetic, 8-bit
Annex G, process 2
1
7
Full progression, Huffman, 8-bit
Annex G, process 3
1
8
Full progression, arithmetic, 8-bit
Annex G, process 4
1
9
Spectral selection only, Huffman, 12-bit
Annex G, process 5
10
Spectral selection only, arithmetic, 12-bit
Annex G, process 6
11
Full progression, Huffman, 12-bit
Annex G, process 7
12
Full progression, arithmetic, 12-bit
Annex G, process 8
13
Lossless, Huffman, 2 through 16 bits
Annex H, process 1
14
Lossless, arithmetic, 2 through 16 bits
Annex H, process 2
Hierarchical mode syntax requires a DHP marker segment that appears before the non-differential frame or frames. It may
include EXP marker segments and differential frames which shall follow the initial non-differential frame. The frame
structure in hierarchical mode is identical to the frame structure in non-hierarchical mode.
Either all non-differential frames within an image shall be coded with DCT-based processes, or all non-differential frames
shall be coded with lossless processes. All frames within an image must use the same entropy coding procedure, either
Huffman or arithmetic, with the exception that non-differential frames coded with the baseline process may occur in the
same image with frames coded with arithmetic coding processes.
If the non-differential frames use DCT-based processes, all differential frames except the final frame for a component shall
use DCT-based processes. The final differential frame for each component may use a differential lossless process.
If the non-differential frames use lossless processes, all differential frames shall use differential lossless processes.
For each of the processes listed in Table J.1, the encoding processes are specified in J.1, and decoding processes are
specified in J.2.
NOTE There is no requirement in this Specification that any encoder or decoder which embodies one of the
above-named processes shall implement the procedures in precisely the manner specified by the flow charts in this annex. It is
necessary only that an encoder or decoder implement the function specified in this annex. The sole criterion for an encoder or decoder
to be considered in compliance with this Specification is that it satisfy the requirements given in clause 6 (for encoders) or clause 7 (for
decoders), as determined by the compliance tests specified in Part 2.
CCITT Rec. T.81 (1992 E)
137