23. Image processing
Theory of Operation Ver. 8.0 Jun. 2008
16
6
C
omposition/Operation
• The following detail the image processing operations performed by MFP board on the
Write section.
1.
FIFO memory is used to enlarge or reduce images in the main scanning direction.
The image is enlarged by increasing the number of data readings and reduced by
decreasing the number of data readings.
2.
Reduction processing is conducted in sub scanning direction. No processing is
done at same size or zoom, but at reduction, the lines are thinned out.
3.
The scanning area is divided into multiple blocks. The ratio of color or monochrome
is calculated for each of these blocks. The machine then determines whether the
entire original is colored or monochrome. A histogram of lightness is also created
for later AE processing.
4.
The scanning area is divided into two in the FD direction. The lightness histogram
for AE processing is generated by the lightness histogram generated through ACS
processing. The AE level of the original is determined based on this histogram and
AE processing is performed.
5.
R, G, and B data are then converted to value and color component data for adjust-
ments of saturation, lightness, and hue.
6.
When copying in a mode other than photo mode, the color component data (R, G,
B) are converted to Y, M, C, and K density data. Also, the masking processing,
which compensates for the deviation in the spectral reflection characteristics of the
toner, and UCR/BP processing are performed on the image data.
7.
Each image area, whether it is a color edge area, black edge area, dot area, or a
continuous gradation area, is discriminated.
8.
Other types of processing performed are the improved reproduction of black text,
edge emphasis, smoothing and color balance.
9.
In photo mode and PC print, each image data of R, G, B is compressed to reduce
the consumption of data capacity.
10. Binary data gets gamma correction at this time. Makes the necessary corrections
so that the printed gradations have linear characteristics, since the image density of
the input image data is not directly proportional to that of the printed image
because of the changing developing and PC drum characteristics.
11. In photo mode and PC print, the image is processed as multi-valued data (8-bit
data). In any mode other than photo, the error diffusion method is employed to pro-
cess the image as binary (1-bit) data.
12. Temporarily stores the BTC-compressed image data and error diffusion data.
13. The stored image data is compressed in the JBIG (Joint bi-level image experts
group) format.
14. When copying in a photo mode or making PC print, the R, G, and B image data are
stored. When coping in a mode other than photo mode, the Y, M, C, and K image
are stored.
15. The image data read from the file memory is uncompressed through a method in a
reverse way of JBIG compression. At this time, image rotating or sorting processing
is conducted.
16. JBIG image data are expanded in the frame memory.
17. When copying in the photo mode or making PC print, each image data of R, G, and
B are stored to reduce data volume. When copying in a mode other than photo
mode, the binary data of Y, M, C, and K are stored.
For PS printing, multi-valued data of Y, M, C, and K are stored.
18. Temporarily stores the image data output from the frame memory.
19. Image data is expanded using the method opposite to that used in BTC compres-
sion. Only when copying in a photo mode (multi-valued data) or making PC pint,
the steps from “19. BTC expansion” to “24. Screening” are performed.
Theory of Operation Y107522-3
