Re: i1 Profiler Scanner module, clipping white
Re: i1 Profiler Scanner module, clipping white
- Subject: Re: i1 Profiler Scanner module, clipping white
- From: Ben Goren <email@hidden>
- Date: Thu, 10 Nov 2016 18:15:49 -0700
On Nov 10, 2016, at 3:30 PM, ygroups <email@hidden> wrote:
> White patch should be 255 one your target
This is incorrect. The D1 white patch on a ColorChecker Passport is typically around L*=96, a*=0, b*=4, which is roughly sRGB R=246, G=244, B=236. An ideal capture device is going to be reporting similar native RGB values.
> If you can’t get it that way from the scanner open the patch with no CM in photoshop and use the levels command to just get it there.
This is very bad advice, unless the intended workflow is to always apply the exact same channel clipping to all images prior to assignment of the native profile -- which itself is an highly questionable practice.
Any half-decent image capture pipeline, whether from a scanner or a camera, is going to offer some method of channel balancing to at least roughly align the neutral axis to the destination space. The consumer ones do it automagically, perhaps with an eyedropper tool to assist the algorithms. Many offer sliders of various kinds to let the user eyeball it. The real stuff lets you specify it explicitly, in addition to whatever user-friendly crutches they might offer.
ICC profiles are only going to be truly valid for the exact same actual scene illuminant and software channel balance settings, though many find them "close enough" for a wider range. But a scanner has a built-in illuminant that should be reasonably stable, meaning that the channel multipliers should be set once and forgotten about, and an ideal ICC profile will be equally valid for all scans made from that device.
Cheers,
b&
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden