Specifications

CHAPTER 2 OPERATION AND TIMING
2-40
COPYRIGHT
©
1999 CANON INC. CANON SUPER G3 FAX BOARD-F1 REV.0 MAY 1999 PRINTED IN JAPAN (IMPRIME AU JAPON)
Reference:
The following is an outline of the progressive bi-level image compression method:
An original image is read at a high resolution. The result is converted to a low resolution, and
the resulting data is coded (compressed) and transmitted. The data is such that, upon reception,
it immediately provides an overall “impression” of the original image, although at a low
resolution.
Data used to improve the resolution of the image contained in the previously sent data follows.
Using this data, the previously sent data (low resolution) is decoded on the receiving side. As
decoding progresses, the initial low-resolution image will taken on a higher resolution.
In practical uses, the receiving side (e.g., CRT screen) can immediately recognize the image,
and can monitor it while the image as the resolution increases. As needed, the receiving side
may suspend reception of the image at any time.
This method of image compression is based on low-resolution images to code high-resolution
images, and requires a page buffer to hold low-resolution images.
2. Single Progression Sequential Bi-Level Image Compression
The following is an outline of the single progression sequential bi-level image compression
method.
While the progressive bi-level image compression method uses multiple layers of resolution on a
page basis (from low-resolution layer to high-resolution layer) for coding and decoding, the single
progression sequential bi-level image compression codes an image from left to right and from top to
bottom in units of horizontal bands (or lines) referred to as “stripes” (hence, the name “sequential”).
This method is called “single progression” because it codes and decodes on the basis of a single
resolution layer.
Reference:
In this method, data is coded in units of stripes, so that it requires a far smaller buffer than a page
buffer.
An image is coded and the resulting coded image data is organized as follows: