PM5 Optical Brightener Correction -> How to achieve in i1Profiler?

[FMueller]

Got this off the X-Rite website.

Emphasis mine. I haven't done any personal testing, wondering if anyone has? Are we all in a a lather about nothing, especially if we aren't choosing papers with high levels of OBA's? Just sayin'...


Date Created: 6/24/2011   Date Modified: 6/24/2011
  Print this document

PM5 Optical Brightener Correction -> How to achieve in i1Profiler?

Question: ProfileMaker 5 did include a software-based option for Optical Brightener Correction (OBC) at least for the Perceptual table in a profile. If the software detected optical brightener in the measurement data, a checkbox for OBC was enabled and could be switched on or off. This helped me to avoid color casts in a profile. How can I achieve this with i1Profiler?

Answer: The new i1Profiler software includes a completely new profiling engine, which provides already by default a most neutral gray axis and this effects on all Rendering Intent tables. This has also been confirmed for papers containing optical brighteners.

Additionally, i1Profiler provides a special option to enlarge the range of how far the neutralisation of grays goes into the more chromatic color areas: Profile Settings -> Perceptual -> Neutralize Gray. This allows to increase the neutrality of the gray axis further.

Owners of an iSis can additionally perform a dedicated Optical Brightener Compensation workflow. In this workflow, a visually based Optical Brightener Compensation will be included in the ICC profiling procedure. This is also a powerful and well-working way, to compensate for color shifts in the gray axis caused by an amount of UV light in a paper or even decide visually on how to build the gray axis. The iSis is capapble to deliver two sets of measurement data – with and without included UV information. Additionally, it comes by default with a set of neutral Gray Cards (produced by Munsell), which are acting as neutral gray references for the visual judgement to identify most neutral grays of the printer.

For dedicated information on how the OBC compensation with iSis is working, click here (-> under construction).

[Ethan_Hansen]

Comparing i1Profiler profiles built from UV-cut vs standard instruments on OBA-laden papers, one can see the effects of the data having an apparent blue cast from the UV-included measurements. Cranking up the Neutral Gray slider exaggerates the yellow cast appearing in printed highlights as the profile "compensates" for the blue paper substrate. This effect is more pronounced on measurements from older, tungsten-illuminated instruments (i1Pro, iCColor, Spectrolino) than the iSis with it's "UV-included" M0 mode.

I'll grant that i1Profiler profiles show less yellow artifacts on OBA-heavy papers than did PMP5 (when Optical Brightener correction was disabled) or MonacoProfiler profiles. The problem is still there, however. Using the iSis OBC module works like a champ. Unfortunately it is designed for folks who only have a single, local printer and paper to measure. If you are working with a press where the time between generating the OBC target and performing the measurements is not immediate, it is difficult to stop and restart the OBC flow.

Using software-based OBA detection and compensation routines achieves almost the same final print quality as does the dedicated iSis OBC module. There are exceptions - we ran into a glossy film that our software choked on but X-Rite's OBC stuff worked like a champ. We use our own OBA correction code, while Argyll and BasICColor Print have similar functionality.

[Jalok]

Cranking up the Neutral Gray slider exaggerates the yellow cast appearing in printed highlights as the profile "compensates" for the blue paper substrate.
...
I'll grant that i1Profiler profiles show less yellow artifacts on OBA-heavy papers than did PMP5 (when Optical Brightener correction was disabled) or MonacoProfiler profiles. The problem is still there, however.
...
Using software-based OBA detection and compensation routines achieves almost the same final print quality as does the dedicated iSis OBC module.
...
We use our own OBA correction code, while Argyll and BasICColor Print have similar functionality.

Curiously, I have the opposite effect with the Neutral Gray slider. All way to the left (zero value) gives me the most yellowish cast with highly OBA-charged papers. When I set it to 100 I get I print more like the PM5 "Correct for Optical Brightener" feature and ArgyllCMS "-f" option. But i1Profiler seems to give us too subtle variation between 0 and 100, so it can't compensate enough fluorescent behavior of OBA-heavy papers. Some months ago I wrote that I would have noticed a high similarity between i1Profiler profiles and those ones created by PM5 with "COB" feature turned on. But more prints showed me I was wrong. In fact, subtle skin and sky tones are easily noticed between one and another profile, and those differences usually destroy our whitebalance finetuning, making it unpredictable.

So, while i1Profiler gives us the most beautiful profiles with mild or no-OBA papers, their yellow cast is unacceptable with b<-8 papers, specially for portraits and other whitebalance-critical jobs. PM5 is not much behind, so I still use it for these problematic papers.

Regarding to your OBA correction code, what does it consist? A software or worksheet that would manipulate measuring data file and correct inside values? Could you share it? Do you sell it? I think this would be the best thing, as someone could apply OBA compensation only when it is really required, preserving no-OBA paper profiles from blue filtering, and not only to perceptual but to every rendering intent.   

[Scott Martin]

To make a long story short if one wants to correct for OBA's in i1Profiler, one should use UVexcluded measurement data or use the OBC profiling method that is only available for certain devices (like the iSis). Just like with Monaco Profiler, the Neutral Gray slider doesn't do anything for RGB profiles.

[Jalok]

To make a long story short if one wants to correct for OBA's in i1Profiler, one should use UVexcluded measurement data or use the OBC profiling method that is only available for certain devices (like the iSis). Just like with Monaco Profiler, the Neutral Gray slider doesn't do anything for RGB profiles.

What do mean with "neutral gray slider does not do anything for RGB profiles"? For me, it does -- erratically for highly OBA-charged papers, and subtle but useful for other papers.

Reading another topic about profiling papers behind glasses, would it be possible to lay above OBA papers a thin UV glass for target measurements? What kind of glass would it need to be?

[Ethan_Hansen]

Jalok: You're correct in the Neutral Gray slider direction - my faulty memory showing. A setting of 0 maps the output to be colorimetrically more neutral while 100 preserves the paper color. This is the opposite operation of MonacoProfilier's Neutral Gray slider.

You are also right in that the Neutral Gray slider affects the Perceptual intent of RGB profiles as well as CMYK. For example, i1Profiler RGB profiles built for a paper with a white b-value of -9.6 (113% reflectance in blue) produce different output values for the same RGB input depending on the setting of the Neutral Gray slider. The difference is subtle for most papers - as with most of the i1Profiler options a 100 point scale is silly. It is not doing the same thing as PMP's optical brightener correction. There is more of an overall color cast rather than compensating for the optical brighteners. The effect is the same as Monaco's Neutral Gray slider or PMP's Paper Colored Gray vs. Neutral Gray option. Both of those options, like the NG slider in i1Profiler, could produce visually curious results on some paper stocks. None of the adjustments are a substitute for OBA compensation.

We licensed a number of algorithms to a certain large software vendor whose operating system-level color management efforts died quietly. We're considering releasing parts of our codebase either as open-source or simple utilities. Selling applications entails providing tech support. That's a losing proposition for products with such a limited market..

Our OBC module might be a good candidate for a standalone tool. It dials in variable corrections based on the measurements. It triggers off both the absolute level of blue tint seen (i.e. b-value of paper white) as well as the shape of the remission curve in the 400-480 nm region. Basically, if enough of a hump is seen, OBA's are present. How much the curve is flattened out depends on both the magnitude and steepness of the spike in blues. Our code is limited, for now, to CGATS files. Supporting X-Rite's mxf files would either take using X-Rite's SDK (and the associated licensing nightmares) or hacking together a customized XML parser.

An alternative, for an industrious person, would be to extract Argyll's OBC code into a standalone utility. It would take adding CGATS file support at a minimum and, perhaps, i1Profiler's mxf file format. We extracted instrument control routines from Argyll for our own use, but aside from a cursory look through the code, I have not looked at it in detail. Even an overview, however, showed that Graeme's code structure of elaborate pointer type-casting to kinda-sorta emulate object-oriented programming in straight C makes matters more difficult for reuse.

[Scott Martin]

What do mean with "neutral gray slider does not do anything for RGB profiles"? For me, it does -- erratically for highly OBA-charged papers, and subtle but useful for other papers.

You're right, it does do a little something if you have super bright white papers. I should have said the results are super slight if visible at all. This slider's effect is more apparent on CMYK profiles. And as Ethan said, it is not a substitute for OBA correction.

Again, to make a long story short, 1) UV excluded measurement data or 2) the OBC profiling method that is only available for certain devices (like the iSis) are the options you have moving forward with i1Profiler.

[neil snape]

Well if the media has little OBA then other than paper white it should not be a radical change?

Above seeing fluorescence makes me think reflections brighter than the emission light source are not necessarily the same thing as OBA reflecting more b channel power.

The exact formula X-Rite (or Gretag Logo) engineers have weighted the expected OBA return values is surely a best guess for the papers used. Fluorescence goes beyond those expectations.