[CinePaint] Really basic script for batch converting

[email protected] Wed, 25 Nov 2009 00:54:03 -0000 (GMT)
Newsgroups gmane.comp.video.cinepaint.user
Message-ID <37179.82.37.43.160.1259110443.R0MUQmxZTEF/[email protected]>
Hi

I've been trying out adding a minute amount of noise to an 16/32bit image
converted from 8bit in Cinepaint as per early Photoshop 15bit + 1bit noise
comments in last email and so far I'm happy with outcome.

I've taken an 8bit image in Cinepaint, convert to 16 or 32bit, add a layer
by duplicating the background, use 'noisify' to add .008 noise to each of
the 3 channels.

Checking the histogram to see if it is all smooth.

Flatten image layers.

Then doing some heavy handed brightness-contrast and HSL I see no gaps
appear in the histogram :-)

I take your advice that gaps are not necessarily a problem.

With lower noise I see gaps appear in the histogram and higher noise value
noise becomes clearly visible in the image.

Adding noise at .008 looks good so far at normal viewing resolution
1920x1080. I can see no visible sign of noise at higher magnification
however yet to view on a HD TV for example.

The image files already suffer from HDV compression blockiness which the
noise appears to disperse a little.

I shoot with no artificial sharpening done in camera and add gentle
amounts of sharpening done with unsharp mask before encoding to video file
once more. Yet to test this with the added noise.

I'm considering GraphicsMagick CLI and using DPX as the format at 16bit
rather than 32bit.

So now looking for a command line script to do:

1.	Convert to 16bit depth
2.	Add a layer by duplicating image background
3.	Add very fine noise to duplicated layer only
4.	Flatten Image
5.	Inverse Gamma to Linear RGB (0.454545)
6.	Assign Linear sRGB editing colour profile
7.	Assign sRGB Display colour profile
8.	Save as DPX with embedded ICC profile if supported.

I've read both GM and IM documentation online but can't find either
duplicate layer or add noise?

Stepping back. With maintaining as much quality as possible from camera
HDV (mpeg2) transport stream files in mind and the YCrCb to RGB conversion
being lossy would it make more sense to go from mpeg2 -> Image format
supporting YCrCb -> Upsample to 4:2:2 (Beneficial??) -> Convert to 16bit
-> Convert to RGB Full Range No remapping -> then continue from 2. above.

Many thanks







------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july