[alt-photo] Print Density Estimation From Scan

[Peter Friedrichsen]

Hi All,

Just wondering if anyone is willing to comment on my method of 
estimating print density using a scanner and a Souffer reflective step wedge.

The density estimate is based on comparing only the red binary values 
with those of a Stouffer step wedge of 0.15 density increments per 
step, and this is done in photoshop by browsing pixels. The red 
channel was chosen because of an iron stain that affects 
predominantly the green and blue channels. The image I used is 
labeled print 3a and can be found here:

<http://picasaweb.google.com/pgfriedrichsen/FerricGumProcess#>http://picasaweb.google.com/pgfriedrichsen/FerricGumProcess# 


The whitest area in the sky on the left side and about 1/2 way up 
gives me a value of 225. The first shades (lightest greys) above the 
left side of the roof give me 221, and the darkest shades just under 
the eaves troughs gives me 42. I have chosen to use the lightest grey 
values because there may be blown highlights and I don't want to 
include those in the density range. So, here is what I am using in summary:

image density-----------pixel value-------step wedge # at this value

lightest greys--------------221---------------less than 2 (use two as 
conservative)
darkest greys---------------42---------------10

difference density=10-2=8

now I multiply 8.0 by the step density difference value of 0.15 per 
step: 8 x 0.15=1.2

I get a density of 1.2.

Anyone wish to comment on whether this is a sound technique for a 
reflective density measurement?

Peter Friedrichsen