ðòïåëôù 


  áòèé÷ 


Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 

  óôáôøé 


  ðåòóïîáìøîïå 


  ðòïçòáííù 



ðéûéôå
ðéóøíá












     áòèé÷ :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: filmscanners: Photoshop Curves vs Levels



If the curve(s) are straight lines, linear, then I guess you should be able
to do the same adjustment in Levels.

This image appears to cry for non-linear adjustment in separate channels. I
don't think the real highlights are in skin tones.

Bob Wright

----- Original Message -----
From: Alan Womack <arwbackup@worldnet.att.net>
To: Majordomo leben.com <Filmscanners@halftone.co.uk>
Sent: Sunday, February 11, 2001 10:41 AM
Subject: filmscanners: Photoshop Curves vs Levels


> Last couple of days I was working on a scan from my ScanWit from Fuji
Super G 400 of my 2 year old son playing with his trucks.  I had used a
couple of flash units and screwed up the light level a bit and certainly the
background shadows.
>
> The local minilab printed crap, but the expression was cute..  So into
VueScan I went, aweful.  Color balance wouldn't come close, way too blue.
Also there was bad posterization in highlights from his face.  I went to
MiraPhoto and found I couldn't control where it placed the actual highlights
what so ever, it was mapping the brightest part (those skin reflections) up
to 255.  Obviously this wouldn't work.
>
> I took a look at a RAW scan from VueScan in PS and found there was a
difference in skin values or maybe 3 RGB and way too much blue.  7-10 points
more blue than red, which is typically in my children 30-40 below red when
skin tones are in the low 200 range.  (the Raw scan was in the 37 to 41
range)
>
> I then took up a curve that was nearly straight up to move my highlights
in tke skin into the  200 neighborhood and spent another couple hours fixing
the skin tones.
>
> So my question is:
>
> Is there a fundamental difference between using levels to stretch the
histogram versus using curves to stretch the same with a very steep line?
>
> alan
>




 




Copyright © Lexa Software, 1996-2009.