Pages: [1]   Go Down

Author Topic: Fixing Epson V850 Scan Errors  (Read 803 times)

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Fixing Epson V850 Scan Errors
« on: July 07, 2018, 05:35:14 PM »

My interest in scanners is quite recent. I anticipated metameric error knowing that current scanners use "white" LEDs that typically have extremely peeked spectra. So I fully expected that profiling the scanner with an IT8 card on Kodak Endura would not yield particularly good scans of a XRite Colorchecker, nor my Epson 9800 prints. I was not disappointed. After profiling to the IT8 card, scanning my XRite Colorchecker card and comparing the dE (All dEs in this post are 1976) values to those measured with an I1Pro2 yielded an average dE of 4.3 across the 24 patches with a max of 8.0 which was due to the last, "black",  mostly fromo a negative a* over -7.

But then why should an IT8 Endura photo print be all that close to a Colorchecker? Besides, the reason I got the scanner was to critically search printer gamut smoothness to evaluate subtle differences between printers not easily measured. This requires searching a significant fraction, perhaps hundreds of thousands of near colors, and is not realistic to do with a spectrophotometer.

So rather than messing around with different IT8 or other photo based targets I made my own, a 1914 patch set printed on 2 pages in the standard used for making printer ICC targets. These were printed, dried, and scanned with the V850 (raw, "no color correction") and Lab values measured with an iSiS. The RGB values were extracted from the scanned patches and combined with the iSiS Lab values and run through Argyll's software, txt2ti3, and colprof, producing an ICC scanner profile.

The profile actually produced slightly (10%) better dEs measuring the XRite CC than the IT8 but still had average dEs in the high 3's. No biggie I thought. Since my hypothesis was that the profile should be well tuned for my 9800 printer that it should be quite good reading CC colors printed from that printer. It was better, with dEs dropping to mid 2's but that was disappointing. Since the scanner target was printed and profiled from 9800 prints, it should produce closer matches as metameric shifts should no longer be a factor. So what gives.

That's when I noticed that a white patch in the scanner profile print, as output by the custom scanner profile, was way off (L* diff of 6) compared to the larger white areas of the scan that were unprinted. Looked at scans using the IT8 profile as well as the V850 defaults. Same differences. It isn't something fixable by profiling because the raw RGB values from the scanner differs. And by a lot.

What could be causing this? So I looked at what the scanner was doing with the lid up and saw two frosted white led light sources on either side of a slot. Deep within the slot is a lens that focuses reflected light coming through the slot.



Examining this while scanning shows intense light surrounding lighter areas leading to the observation that light was being reflected off the paper's brighter areas, bouncing off the frosted white surfaces of the LED covers, and adding back into the light illuminating the area being scanned.

Next, I measured the effect by scanning increasing diameters of white circles embedded in black. A large, 50mm, white circle produced about 22% more light than small circles of 3mm and 18% more light than a 6mm square (which is the default iSis and IT8 patch size). Below 2mm, diffraction effects start to dominate while the light reflected from a 2mm circle and bouncing back drops below a 1% effect.

Next I measured horizontal and vertical reflected light with very large squares of white/black and measuring the reflected light increase as distance from the black squares increased. I also measured the change at a diagonal to get a feel for how reflected light from a vertical and horizontal offset combined. I also made a randomized matrix of small, 4mm, black and white patches and did a regression analysis to determine the reflected light at offsets from 4mm white patches as a cross check.

Then I used a curve fit to the horizontal and vertical offsets to create a function that estimated reflected light given an x,y position offset. This function was then applied to the scanner image, pixel by pixel, within about 25mm, to estimate the light, including color, that was reflected from other nearby areas.

This estimate of reflected light, in linear space, was then subtracted from the scanned RGB values, also in linear space and the result was written to a new tif file.

Now I created a page, in iSis scannable format, of 5 high tone neutrals from L*90 to L*98 in Rel. Col. These were embedded in a white surround and also a black surround. I did the same for the 24 CC patches as read from my "Classic" Colorchecker but using Abs. Col. Again, in a white surround and black surround. This was measured with an Isis and scanned with the V850. Here is what the scan looks like:





Looking at just the high tone patches in the white and black surround, we get these results:

First, shown with the L*a*b* values as measured by the iSiS:



Next, as scanned and profiled normally:



Finally, as scanned with the scan corrected for reflected light:



I haven't show the detail for the Colorchecker patches but the improvements are similarly dramatic.

  L90 to L98: Ave. dE between white and black backgrounds
  |      ColorChecker: Ave. dE between white and black backgrounds
  |      |
 6.30   4.67  Printer based Argyll profile
 0.87   0.86  Printer based Argyll profile, Refl. removed


Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 11621
    • http://www.markdsegal.com
Re: Fixing Epson V850 Scan Errors
« Reply #1 on: July 07, 2018, 08:19:12 PM »

Nice detailed analysis, and you are indeed onto something real.

The problem of scanner flare has been known for quite some time and isn't confined to the Epson V850. Don Hutcheson discussed it in some depth in his Scanning Manual, May 2011 (Optical Flare page 21) and recommended masking around the media with black surround material. Well-designed media frames also help. My own research has indicated the problem is worse the glossier the media (stands to reason). The black surround solution is only partially useful when large parts of the image are bright and help light to bounce around where you least want it. I have also determined that profiling quality can make a difference. This, amongst many other subjects, will be apparent in my forthcoming article on scanner profiling.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Re: Fixing Epson V850 Scan Errors
« Reply #2 on: July 07, 2018, 10:35:10 PM »

Nice detailed analysis, and you are indeed onto something real.

The problem of scanner flare has been known for quite some time and isn't confined to the Epson V850. Don Hutcheson discussed it in some depth in his Scanning Manual, May 2011 (Optical Flare page 21) and recommended masking around the media with black surround material. Well-designed media frames also help. My own research has indicated the problem is worse the glossier the media (stands to reason). The black surround solution is only partially useful when large parts of the image are bright and help light to bounce around where you least want it. I have also determined that profiling quality can make a difference. This, amongst many other subjects, will be apparent in my forthcoming article on scanner profiling.
Mark, most of Don's scanner flare discussion is related to scanning transparencies. Especially 35mm slides. This is a different beast and much harder to deal with. The suggestions to thoroughly clean the glass and mechanism are dead on but make little difference for reflective media scans where the flare can be as high as 20%.  That's beyond a bit of dust. However, dealing with transparency flair is quite hard and he well describes the issues.

The V850 flair is shockingly high, with reflected light from larger white areas causing 20% more illumination. Interestingly, the way human vision works it's not a very noticeable effect perceptually. It's very hard to see that a light patch, surrounded by darker colors, is really darker than the same white patch surrounded by a large area of light colors. It plays right into an optical illusion. Off white patches surrounded by dark colors look whiter than they are so if they are actually scanned as darker the visual effects are barely noticeable. It's only when one starts to measure things.

But the good side of the V850 reflected light flare errors is that they are almost entirely predictable. This is not true of transmitted light flare where a mere 1% additional light can result in scans that are visually off. And, whether 1% or 2% or even .5%, it is extremely hard to control and is less deterministic.

But I digress. My interest is mostly on colorimetricly accurate, reflected light scanning and profiling. The very large reflected light flare doesn't find it's way to the scanner lens without first illuminating the scanned portion of the target. And it's almost entirely a function of the amount of light reflected from a print/photo over about a 1 inch area around the focal point. And this can be calculated quite accurately and removed from the image.

My testing shows that there is some difference between glossy, semigloss, and matte reflective scans but they are very small in comparison to the large (20%) flare from a large, bright area.

However, I think small improvements can be made in my methods. An assumption that has worked well so far is that the light reflected from a scanned print area is attenuated evenly for each of the 3 channels. This is likely not the case. just like light reflected off a sheet of paper will have a slightly different CCT than the incident light because few reflectors are truly neutral. Also, reflected light may not be attenuated at the same levels a spectro would indicate due to variations in bronzing, etc. from incident light at low angles. But these are second order effects and likely vary a bit depending on printer inks and such.

In any case the corrections my model makes reduces the flare error at higher light levels to the point that the major high luminance residual errors are due to metameric effects from white led's compared pr, presumably, D50 and the different spectral characteristics of inks, photo chemistry, and such.

Low luminance errors continue to be due to dust, imperfections, and such which are well known as they tend to be visually apparent.

Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 11621
    • http://www.markdsegal.com
Re: Fixing Epson V850 Scan Errors
« Reply #3 on: July 08, 2018, 09:01:46 AM »

This is very interesting and useful Doug. It could give the software developers some ideas on building such custom improvements into their algorithms - the problem being of course that it would vary by scanner and media, so perhaps quite a job.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Re: Fixing Epson V850 Scan Errors
« Reply #4 on: July 08, 2018, 11:28:01 AM »

This is very interesting and useful Doug. It could give the software developers some ideas on building such custom improvements into their algorithms - the problem being of course that it would vary by scanner and media, so perhaps quite a job.

It may, in fact, be quite very consistent between media types. I've now tested this on matte, glossy, and lustre. Also on very high OBA and no OBA papers and see very little difference. I was particularly concerned about OBA papers because the reflected light would differ from the incident light. Reflected light from OBA papers contains little uV as it's converted to visible light. But it turns out the uV content of the "white" LEDs is effectively zero so there is no impact from OBAs. Of course this depends on the scanners using similar LEDs.

But yes, it would vary depending on the scanner design so I would expect differences amongst scanners. It's possible to design the scanner illumination w/o frosted surfaces but that also would require more LEDs to achieve the required evenness of illumination since the frosted surfaces serve to more evenly distribute light.

However, for a particular scanner design it should be quantifiable and reversible by software. At least for those that use illuminants w/o uV.

Correcting scanned images for reflection also improves profile statistics since some of the error in calculating profiles comes from variations in reflected light from other nearby patches. However, since most nearby patches are not very luminous (average L* of 50 means only around 20% of incident light is reflected and so the additional light a patch sees is .20*.20 or about .04 which will affect the typical scanned patch by only a bit over 1dE.

There is a term for this, "Large area spatial crosstalk," and a means of somewhat quantifying it in IEC 61966-8. Section 13.1-13.4 provide details. As such it could be specified by scanner manufacturers but I don't see any evidence any do so based on cursory searches.
https://webstore.iec.ch/preview/info_iec61966-8%7Bed1.0%7Den.pdf

I suspect that this gets little attention because the effect is not normally visually noticeable. While colorimetrically, the effect can be quite large (20% increase in reflected light in large, near white areas) it occurs in a way that it just isn't a material visual effect. At least for most scanned images.

But for those desiring a good colorimetric result it's a major effect. It's possible to use a scanner as a poor man's densitometer and the reflected light would be a significant factor. Corrected for reflected light goes a long way towards improving a scanner's use for such purposes.

My focus on this was from trying to get the best, colorimetrically accurate, scan of prints in order to illustrate issues around printer imperfections. For that purpose, the effect is material but fortunately, correctable.
Logged

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Re: Fixing Epson V850 Scan Errors
« Reply #5 on: July 08, 2018, 01:06:08 PM »

It's worth discussing the general issue of flare. Almost everyone here is familiar with camera flare and how it varies with internal reflections off of lens elements and housing. Sometimes to a nice artistic effect. But also the general issue of flare reducing dynamic range. User sources of flare include things like smudges on the lenses and general environmental dust or particulate accumulation as well as reflections intrinsic to glass/air interfaces.

However, it is very rare for camera/lens flare to exceed 1% and fractions of that are more typical.

Scanners, like cameras, also suffer from this sort of flare. The scanner's lens structure, as well dirt, fibers, and even imperfections in the scanner glass, all contribute. This, like in cameras, reduces the scanner's ability to discern small changes in deep shadows. And these effects can be very visually noticeable.

The large area crosstalk scanners suffer from is most curious. It is simultaneously, far larger than the "regular" flare that afflicts both cameras and scanners, but also is a kind of flare that is rarely visually noticeable. Further, correcting for large area crosstalk flare does nothing to reduce the impact of "normal" flare in darker areas of a photo. It can even increase it. This is because large area flare serves to increase contrast and boost brighter areas of a scanned image. And that can make the regular flare less visible since it is not enhanced by the contrast boost.


One trick I found very useful for inspecting the scanner glass surface is to use a bright (250 or more luments)  LED flashlight with a tight beam. Shining it through the glass at an angle shows even tiny bits of smudge. I found residual ink patterns from prints that had dried for days were left on the glass even though it was completely invisible w/o the LED inspection. But if I scanned a black print and boosted visibility with curves in Photoshop those same residual areas show up. These required a good glass cleaner to remove.

Since discovering this I've started baking my prints at 60C for an hour. This seems to work to get rid of the glycol part of the ink that was leaving the tracks.  The glycols have a very low vapor pressure and stick around a long time without this process. Not recommended normally as it causes a 1 to 2 dE change in colors compared to letting prints dry 48 hours but it does speed up getting the glycols off so that I don't recontaminate my scanner glass and is useful for most experimental processes.
« Last Edit: July 08, 2018, 01:15:14 PM by Doug Gray »
Logged

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Fixing Epson V850 Scan Errors - Program and sample file included
« Reply #6 on: July 08, 2018, 05:25:18 PM »

I've included an executable program which removes reflection from an Epson V850 type scanner and should work with the V800 as well as the scanner module is the same. The program is named scannerreflfix.exe and includes several DLLs needed for decoding tif files. It's usage is:

Usage: scannerreflfix [-I] [-W] -[P ProfileFileName] [-S EdgeRefl] infile.tif outfile.tif
  -I                      Save intermediate files
  -W                    Adjust for white (Relative Col with tint retention)
  -P profile           Attach profile <profile.icc>
  -S edge refl;      average reflectance of areas outside of scanned area (0 to 1, typ .85)

The "I" option saves various intermediate image files used for estimating and removing reflected light. The most interesting one is refl_light.tif which is the additional extra light reflected back and forth from the image to the frosted structure and is removed from the processed image.

The "W" option adjusts the brightness upwards while retaining colorimetric accuracy. It is similar to Relative Colorimetric conversion except that tint is not changed. In true Rel. Col., the white point is set so that paper tint becomes white. Not typically desirable for scanners. This most closely simulates the higher brightness that the reflected light produces as it tends to lighten the white borders around an image.

The "S edge refl" option is used to set the reflectance of the unscanned portion. For instance stamps are often scanned with a black background or without closing the lid/cover. In such as case use "-S 0" so the program will estimate reflected light correctly. The default of .85 works pretty well for scanning against a white background or the flat cover provided.

The "-P profile" will embed an ICC profile into the processed image file. I have included the Argyll profile made from a 1914 patch set on my 9800. After processing, image colors average under 1.0 dE00 over a wide range of prints made with the 9800. I have found that even though it was made from 9800 printed colors and not an IT8 Kodak Endura, scanning an actual Colorchecker produces slightly better results than using the provided IT8 and a profile generated from I1Profiler. Average dE was a bit over 2 for the 24 XRite Colorchecker patches.

Also provided is a set of 4 squares with small, L*85 patches, which can be processed by the scannerreflfix program. A command file, "example.bat" is provided which runs it.

Here are images of the files before and after fixup:


BEFORE



AFTER


And here is the program's calculated reflectance. That is, the light reflected from the scanned sheet to the frosted surfaces on the scanner LED covers, then back to the image. The reflected light is highly influenced by nearby colors as can be see from the amount of yellow being reflected on the first patch.




The DeltaE values between the BEFORE and AFTER neutral patches are, from left to right:

8.1
3.4
3.1
4.5

After the correction is run, the dEs are within +/- 1 each other as well as those printed and measured with a spectro.

The program has been tested on OBA free matte, high OBA glossy, and low OBA luster and the reflection correction works quite consistently with each.
« Last Edit: July 08, 2018, 06:00:05 PM by Doug Gray »
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 11621
    • http://www.markdsegal.com
Re: Fixing Epson V850 Scan Errors
« Reply #7 on: July 08, 2018, 05:45:02 PM »

Neat, but I assume it being ".exe" this is Windows only?

:-)
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Re: Fixing Epson V850 Scan Errors
« Reply #8 on: July 08, 2018, 06:22:46 PM »

Neat, but I assume it being ".exe" this is Windows only?

:-)

Yep. Source is in portable C++ but needs the tifflib for encoding/decoding tif images. Probably pretty easy for someone with a Mac to port as there is nothing Windows specific in my code. Any volunteers?
Logged

Eric Myrvaagnes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 15225
  • http://myrvaagnes.com
    • http://myrvaagnes.com
Re: Fixing Epson V850 Scan Errors
« Reply #9 on: July 08, 2018, 08:18:09 PM »

Thank you Doug and Mark for this fascinating and useful discussion.
I am delighted to hear that your "fix" program applies also to my V800 scanner.

I have learned a lot from this thread.

-Eric
Logged
-Eric Myrvaagnes (visit my website: http://myrvaagnes.com)

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 11621
    • http://www.markdsegal.com
Re: Fixing Epson V850 Scan Errors
« Reply #10 on: July 08, 2018, 08:56:14 PM »

Now that Eric mentions the V800, I should say it's been running through my mind that maybe the same or similar problem and solution could also apply to the V750/700 models as well, insofar as the basic design is the same. The big change was the conversion from CFL to LED illumination - I don't know how much a difference that makes in this regard.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Re: Fixing Epson V850 Scan Errors
« Reply #11 on: July 10, 2018, 12:13:12 AM »

Now that Eric mentions the V800, I should say it's been running through my mind that maybe the same or similar problem and solution could also apply to the V750/700 models as well, insofar as the basic design is the same. The big change was the conversion from CFL to LED illumination - I don't know how much a difference that makes in this regard.

Good chance it will work on CFL lit scanners too. If you're game, we can find out and I'll post a file to print. Just scan it and post the tif file. You will probably have to load it in Photoshop and save as with compression for it to fit the forum's limits. Would be nice to get a scan of a high OBA and low OBA print to see if there are uV related issues though, at most, it's a secondary effect. My 850 doesn't have significant uV so it would be interesting to see if there are any effects from OBAs on a CFL scanner.

As an aside, I printed and scanned a reference image, the Kodak PDI, and the uncorrected image has a noticeable increase in luminance in the brighter areas together with a slight reduction in the luminance of the darker areas compared to both the corrected image and the original when toggling back and forth in Photoshop.
« Last Edit: July 10, 2018, 12:17:24 AM by Doug Gray »
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 11621
    • http://www.markdsegal.com
Re: Fixing Epson V850 Scan Errors
« Reply #12 on: July 10, 2018, 07:46:21 AM »

I don't have a V750 any longer, only the V850, so I can't test it.

On the V850 I did test lower versus higher OBA paper and determined worse results with light tones (L*70-94) in the higher OBA paper, but that paper is also high gloss, whereas the other is luster, so it's quite possible that the high gloss would be more influential for reflectance/scatter issues than the OBA content. But that's a presumption, I'd need a low-gloss high OBA paper to verify it, and won't be doing that any time soon - it's gilding the lily for my purposes.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1381
Re: Fixing Epson V850 Scan Errors
« Reply #13 on: August 16, 2018, 03:16:35 PM »

Most of my work trying to improve scanner colorimetric accuracy is to detect areas where the printer isn't producing exactly the right colors. Also, detecting areas where the printed gradients show shifts that aren't smooth. While my i1iSis can read through a large patch set reasonably quickly, it isn't practical to produce patch sets in the millions which would be needed to find subtle, printer non-linearities.

But on to the scanner: I printed a 4x5, using perceptual intent, at 720dpi, an image from a trip to the Trinity site at the White Sands base. IIRC, the kids are playing inside a broken large cylinder that had been used in a TNT calibration test a bit before the actual Trinity test.

This was scanned using various profiles and the results were overlayed in Tiff file layers.

The layers are:
Photoshop Proof
scanned, IT8 Target and I1Profiler
scanned, built in default Epson ICC profile
scanned, 2871 pch printed target, Argyll profile
scanned, 2871 pch printed target, Argyll profile, Refl. corrected.

The prints were made with a printer profile made from the same printed patches that were used to create the two Argyll profiles.

The Photoshop "proof" layer was made by converting the image to printer device space then converting to ProPhoto using Abs. Col.  This is equivalent to setting "view proof" to the printer profile and selecting Perceptual and "Show paper color" in Photoshop.

Printed and scanned at 720 DPI. Profiles were applied to scans which were printed and scanned with a white surround but trimmed to 4x5. Images were reduced to 180DPI to keep the file size reasonable. Color shifts from downsizing are negligible.

If the image is downloaded and viewed in Photoshop you will see these 5 layers and, by selecting the "proof" layer and toggling on and off the scanned layers one can see even slight imperfections.

Here's what I see relative to the "proof" layer:

All the profile layers, except the reflection corrected on, show a relative increase in contrast with the near white areas of the sky increasing in luminance compared to the inside of the large cylinder.

The IT8 created profile and Epson's built-in ICC profile scans show a slight shift to the blue in the high key areas. The differences between the IT8 and Epson's canned profile are less than the differences to the "proof" layer.

The reflection corrected scan using the 2871 printed patch, scanner profile shows the least change. It doesn't show the contrast increase the other scans do. This is expected since the contrast increase occurs because reflected light from nearby high key areas increases the reflected light at the scanned point.

However, there is a slight shift in the brown (inside of cylinder) and orange (bucket on table) with the colors becoming slight more saturated in the scans compared to the "proof"

So this brought up the question: Did the printer accurately produce the color it think it did? If so the Lab values from the "proof" layer should be close to measured colors on the print.

OTOH, if the measured, printed colors matched the "proof" layer then the scans contained most, or all, of the error.

I measured the average colors in both locations and the scans were more accurate than the "proof" and both the prints and scans were about 5 dE (1976) more saturated.

This is quite a bit since my printer is highly consistent printing colors accurately. Most colors are under 1 dE and only rarely do they exceed 2. So what is going on?

In examining these two areas, there is one major trait. Both the orange bucket and inside of the cylinder have a considerable amount of fine grain texture. This is much more obvious on the 720DPI originals. The orange bucket actually has horizontal grooves. It seems likely that the printer simply does not reproduce this fine spacial texture as accurately as it does the solid, fixed RGB colors used to create profiles.

This bears further investigation. I think I'll measure the image texture noise statistics and create textured i1iSis patch sets with specific averages and see if the printer's color shifts as a result. If someone wishes to do a test like that I suggest keeping the texture fine enough that the average over an area of a square mm are reasonably deterministic. The XYZ values of each pixel should be calculated and averaged over that area then turned back into Lab as Lab is highly non-linear.

Update: I just printed patches with and without high spatial freq variations in the colors of the orange bucket and brown inside cylinder such that the XYZ values averaged the same as solid patches. The printed patches, both solid colors and the ones with spatial variation measured the same within fractions of a dE. And both were different from what the profiles indicated should have been printed and were consistent with the variation and direction of change observed between the scans and the image "proof."  It appears the printer's profile is in error in these places. Interesting in that the average dE00 is about .6 with that profile based on prior tests. OTOH, I did open a new box of print paper. Time to redo and verify the printer profile.

I was frankly surprised at how consistent the spatial variation colors read. It was not what I was expecting. The printer does a good job making it's ink drop pattern to do that. At this point the scanner profile and reflection correction gives better colorimetric readings than the printer/profile.




link to tif:
https://1drv.ms/u/s!AoQDySMc0uNihf8FJU9ysqUAHN5lJA

« Last Edit: August 17, 2018, 12:48:14 AM by Doug Gray »
Logged
Pages: [1]   Go Up