FW: **2-D Calibration Can someone shed light on this Xerox announ cement?
FW: **2-D Calibration Can someone shed light on this Xerox announ cement?
- Subject: FW: **2-D Calibration Can someone shed light on this Xerox announ cement?
- From: "Nuuja, Robert" <email@hidden>
- Date: Fri, 21 Nov 2003 09:54:23 -0500
-----Original Message-----
From: Bala, Raja
Sent: Friday, November 21, 2003 9:47 AM
To: Nuuja, Robert; Rolleston, Rob J; Bala, Raja
Subject: RE:Can someone shed light on this Xerox announcement?
Rob,
this was only an oral presentation at the Color Imaging Conference. There
was no paper - however an abstract was published. I have included it below,
I think it will serve to respond to Roberto's basic questions. You can
forward it, or have me respond directly. In the interest of time, I prefer
the former. Thx, Raja..
email@hidden
http://chester.xerox.com/~raja
---------------
Two-dimensional transforms for device color calibration
Raja Bala, Gaurav Sharma, Vishal Monga
IS&T/SID 11th Color Imaging Conference, Nov 4-7, 2003.
Abstract
Achieving consistent, high quality color reproduction in a color imaging
system necessitates accurate color correction of the devices in the system.
For output devices, color correction can be parsed into two steps: 1)
calibration, which maintains the device with a specific color response; and
2) characterization, which establishes the relationship between device
dependent and device independent color for a calibrated device.
This presentation focuses on output device calibration, with emphasis on
printers. Printer calibration is often performed using one-dimensional
channel-wise tone-response corrections (TRCs). While 1-D TRCs are attractive
for their computational simplicity, their use severely restricts the extent
of control that can be exercised within the device gamut. A common example
is that TRCs linearizing each of C, M, Y independently will not result in
gray-balance (i.e. a*=b*=0) along the C=M=Y axis, and vice versa. The
alternative is to use 3D (or 4D) transforms to achieve full control within
the device gamut. These however incur high computational and measurement
effort.
This presentation proposes a novel use of 2-dimensional transforms for
printer calibration. The transforms are implemented as full-resolution 2-D
lookup tables that require no interpolation. This enables significantly more
control over the device color gamut when compared to 1-D transforms, while
incurring considerably lower computational and measurement cost than 3-D or
4-D transforms. Results show significant improvement in calibration accuracy
and stability when compared to traditional 1-D calibration. The 2-D
framework can also be extended to encompass softcopy device calibration.
---------------
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.