Articles: Other

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 ]


In this section I will compare the colors of the original IT8.7/2 target with the colors produced after the digitization process. Finding such discrepancies is one of the functions of the Color Managing System whose main purpose is creating a profile (here, the scanner profile) to use it later to digitize standard originals. Using the numbers for each of the colors of the target, I construct a diagram with the scanner’s gamut range and calculate its area. I won’t describe this process at length – you can refer to my earlier reviews for details.

The gamut range identified as a triangle in the diagrams is a projection on the plane. It outlines the scanner’s palette in the visible light spectrum. Since we know the coordinates of the vertexes of the triangles, we can calculate their areas to estimate the absolute values of the ranges.

To give you some reference points the diagram above also contains the areas of the analogous triangles for the standard sRGB and Wide Gamut RGB color profiles (the former characterizes the gamut range of the majority of monitors, and the latter is the maximum area of the CIE La*b* space).

To build the next diagram I found the difference between the colors of the original target and the scan. The Color Difference value should not exceed 5.0 or the color discrepancies become visible for a majority of people.

The closer to X axis – the better

The results of the Perfection 4180 are shown in the diagram for the comparison’s sake. This scanner has the same optical resolution but a narrower optical density range (3.4Dmax). It can be easily seen that the Perfection 4180’s errors in reproducing colors fall on the darkest areas of the target, starting from the GS17 point, while the Perfection 4990 easily distinguishes between those zones. So, the specified 4.0Dmax is confirmed by the results of my tests.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 ]


Comments currently: 3
Discussion started: 12/21/05 11:49:02 AM
Latest comment: 02/04/07 01:34:37 AM

View comments

Add your Comment