Page 169 - ITU Journal, ICT Discoveries, Volume 3, No. 1, June 2020 Special issue: The future of video and immersive media

ITU Journal: ICT Discoveries, Vol. 3(1), June 2020

Fig. 14 – The situation of the still image (SI) standard within the ITU-T T.120 [13] system architecture
4.1.5 Open document architecture (ODA) raster graphics content architecture: ITU-T T.417 |
ISO/IEC 8613-7 [21]

Amendment 1 to ITU-T T.417 | ISO/IEC 8613-7 [21] supports the use of joint bi-level image experts group
(JBIG) and JPEG colour in applications using raster graphics content,
“The JPEG encoding schemes defined in CCITT Rec. T.81 | ISO/IEC 10918-1 specify two classes of coding
processes: lossy (not information preserving) and lossless (information preserving). The lossy procedures
are all based on the Discrete Cosine Transform (DCT) and the lossless are based on a predictive technique.
Four modes of encoding are defined: the sequential DCT-based mode, the progressive DCT-based mode, the
sequential lossless mode, and the hierarchical mode.
In the sequential DCT-based mode 8 × 8 blocks of pixels are transformed. The resulting coefficients are
quantized and then entropy coded (losslessly) by Huffman or arithmetic coding. The pixel blocks are
typically formed by scanning the image (or image component) from left to right, and then block row by
block row from top to bottom. The allowed sample precisions are 8 and 12 bits per component sample. Of
the DCT-based methods, the sequential DCT-based mode requires the least amount of storage as a file.

For the progressive DCT-based mode, the quantized coefficients for the complete image component are
determined, stored, and processed by either spectral selection or successive approximation. These two
techniques may be used separately or may be combined in various ways.

© International Telecommunication Union, 2020 147