Comparing L*a*b* measurement scales: Photoshop vs Xrite
I have a set of L*a*b* measurements of color charts that I created with a i1Pro2 spectrophotometer. The Xrite software exports L*a*b* values on a scale of L*:0-100 and *a*b*: -100 to +100*. I've written a script for Photoshop to automate the calculation of specific metrics of captures of those charts, but there's one snag: Photoshop displays L*a*b* values on a scale of L*:0-100 and *a*b*: -128 to +127*. I assume, but I haven't been able to find confirmation, that it is a mathematically linear conversion (mapping) from the *-100 to +100 a*b* scale* to the *-128 to +127 a*b* scale.* Can anyone verify this? Or point me in the direction of the proper way to map the values between the two scales? Thanks, Mike -- *Mike Pucher* *Sr. Image Capture Specialist* *Animation Research Library* desk: 818.544.4185 (8223-4185)
i use: double inv_map_ab(double ab) { return (ab+128)*256/UINT16_MAX; } interested to know your results: please let me know if it works for you! regards, edward
On Apr 9, 2015, at 2:26 PM, Mike Pucher <michael.pucher@disneyanimation.com> wrote:
I have a set of L*a*b* measurements of color charts that I created with a i1Pro2 spectrophotometer. The Xrite software exports L*a*b* values on a scale of L*:0-100 and *a*b*: -100 to +100*.
I've written a script for Photoshop to automate the calculation of specific metrics of captures of those charts, but there's one snag: Photoshop displays L*a*b* values on a scale of L*:0-100 and *a*b*: -128 to +127*.
I assume, but I haven't been able to find confirmation, that it is a mathematically linear conversion (mapping) from the *-100 to +100 a*b* scale* to the *-128 to +127 a*b* scale.* Can anyone verify this? Or point me in the direction of the proper way to map the values between the two scales?
Thanks, Mike
-- *Mike Pucher* *Sr. Image Capture Specialist* *Animation Research Library*
desk: 818.544.4185 (8223-4185) _______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/etaffel%40me.com
This email sent to etaffel@me.com
Am 09.04.2015 um 20:26 schrieb Mike Pucher:
I have a set of L*a*b* measurements of color charts that I created with a i1Pro2 spectrophotometer. The Xrite software exports L*a*b* values on a scale of L*:0-100 and *a*b*: -100 to +100*.
Is that really the case? It would surprise me, because the range of a* b* is not limited, and even less likely scaled.
I've written a script for Photoshop to automate the calculation of specific metrics of captures of those charts, but there's one snag: Photoshop displays L*a*b* values on a scale of L*:0-100 and *a*b*: -128 to +127*.
This is due to the ICC specification which defines an encodable range of a* and b* of -128 to 127. Values outside that range are not encodable in an ICC context, and I guess that's why it's like that in Photoshop. Florian.
On Apr 9, 2015, at 11:26 AM, Mike Pucher <michael.pucher@disneyanimation.com> wrote:
I have a set of L*a*b* measurements of color charts that I created with a i1Pro2 spectrophotometer. The Xrite software exports L*a*b* values on a scale of L*:0-100 and *a*b*: -100 to +100*.
I've written a script for Photoshop to automate the calculation of specific metrics of captures of those charts, but there's one snag: Photoshop displays L*a*b* values on a scale of L*:0-100 and *a*b*: -128 to +127*.
I assume, but I haven't been able to find confirmation, that it is a mathematically linear conversion (mapping) from the *-100 to +100 a*b* scale* to the *-128 to +127 a*b* scale.* Can anyone verify this? Or point me in the direction of the proper way to map the values between the two scales?
Hi Mike, How do you know the range from the “X-Rite Software” is -100 to +100 for a* and b*? I see no reason to believe that you should scale anything. I would think you are safer thinking of your -100->100 data as simply range limited and you should encode it to the same values in Photoshop regards, Steve
Thanks for the input, everyone. It is clear to me now that I was confused by the data export options in i1Profiler, and that no correction to the data is needed. I misunderstood the output scale of "0.0-1.0" or "0.0-100.0" as a proprietary handing for L*a*b*, when that option only applies to exporting spectral data. (Learned that from Robin off the list, can't take any credit for figuring that one out on my own.) On Mon, Apr 13, 2015 at 10:56 AM, Steve Upton <upton@chromix.com> wrote:
On Apr 9, 2015, at 11:26 AM, Mike Pucher < michael.pucher@disneyanimation.com> wrote:
I have a set of L*a*b* measurements of color charts that I created with a i1Pro2 spectrophotometer. The Xrite software exports L*a*b* values on a scale of L*:0-100 and *a*b*: -100 to +100*.
I've written a script for Photoshop to automate the calculation of specific metrics of captures of those charts, but there's one snag: Photoshop displays L*a*b* values on a scale of L*:0-100 and *a*b*: -128 to +127*.
I assume, but I haven't been able to find confirmation, that it is a mathematically linear conversion (mapping) from the *-100 to +100 a*b* scale* to the *-128 to +127 a*b* scale.* Can anyone verify this? Or point me in the direction of the proper way to map the values between the two scales?
Hi Mike,
How do you know the range from the “X-Rite Software” is -100 to +100 for a* and b*?
I see no reason to believe that you should scale anything. I would think you are safer thinking of your -100->100 data as simply range limited and you should encode it to the same values in Photoshop
regards,
Steve
_______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/michael.pucher%40dis...
This email sent to michael.pucher@disneyanimation.com
-- *Mike Pucher* *Sr. Image Capture Specialist* *Animation Research Library* desk: 818.544.4185 (8223-4185) mobile: 310.977.3801 street: 1400 Flower St. Glendale, CA 91201-8265
participants (4)
-
edward taffel
-
Florian Höch
-
Mike Pucher
-
Steve Upton