Greyscale transform

DW
Posted By
David Williamson
Jul 6, 2004
Views
491
Replies
3
Status
Closed
I have 2 greyscale images which join, however the range of greyscales do not match between images, how do I transform a greyscale image so it nicely matches the joining greyscale image, I have tried a histogram transform using a slew using the images mean and standard deviation of its luminace in LAB color mode with minimal result.

Any ideas appreciated

Master Retouching Hair

Learn how to rescue details, remove flyaways, add volume, and enhance the definition of hair in any photo. We break down every tool and technique in Photoshop to get picture-perfect hair, every time.

T
tacitr
Jul 6, 2004
I have 2 greyscale images which join, however the range of greyscales do not
match between images, how do I transform a greyscale image so it nicely matches the joining greyscale image…

Easy. Image->Adjust->Curves.


Art, literature, shareware, polyamory, kink, and more:
http://www.xeromag.com/franklin.html
MR
Mike Russell
Jul 7, 2004
David Williamson wrote:
I have 2 greyscale images which join, however the range of greyscales do not match between images, how do I transform a greyscale image so it nicely matches the joining greyscale image, I have tried a histogram transform using a slew using the images mean and standard deviation of its luminace in LAB color mode with minimal result.
Any ideas appreciated

Method 1
———-
If you have CS or Elements, try the Photomerge command. It may do the job right off the bat.

Method 2
———-
The simplest manual procedure is probably to join your images as separate layers, drag them into position, then add a curves adjustment layer and group it with the top layer.

Then adjust the black and white endpoints of the curve to get the best match possible. Add a middle point of the curve, moving it up and down to match the overall brightness of the image.

Three points will probably do it. If that is not accurate enough, add two more curve points, midway between the other points. I doubt you will need more than this, but you may of course continue adding curve control points until you no longer see any improvement in the match between the two images. This procedure is less satisfactory for color images, but you may try doing the procedure once for each channel.

Method 4
———-
Here’s a full blown procedure, that I think is overkill for a grayscale image, but it is very precise and will work well for color images:

1) create two or three eyedropper points on each image, in the critical areas that need to match. For example, quartertone, midtone, and three-quartertone. Sky is usually a problem for outdoor shots, so you may want to include two different points in the sky, say for cloud and clear sky.

2) write down the values for the image you are happy with, then use curves to make those points match in the other image.

Method 5
———-
Curvemeister offers a more efficient refinement of method 4, using a process called "Curve Extraction". You may create your eyedroppers in the first image, lock them to their colors, save them to a file, load them into your second image, and fine tune the eyedropper postiions on the second image. You may try this procedure out using only the demo, save the resulting .acv file, and apply the curve in Photoshop’s curves.


Mike Russell
www.curvemeister.com
www.geigy.2y.net
J
Jittered
Jul 12, 2004
Thanks for the help, much appreciated

David

"Mike Russell" wrote in message
David Williamson wrote:
I have 2 greyscale images which join, however the range of greyscales do not match between images, how do I transform a greyscale image so it nicely matches the joining greyscale image, I have tried a histogram transform using a slew using the images mean and standard deviation of its luminace in LAB color mode with minimal result.
Any ideas appreciated

Method 1
———-
If you have CS or Elements, try the Photomerge command. It may do the job right off the bat.

Method 2
———-
The simplest manual procedure is probably to join your images as separate layers, drag them into position, then add a curves adjustment layer and group it with the top layer.

Then adjust the black and white endpoints of the curve to get the best
match
possible. Add a middle point of the curve, moving it up and down to match the overall brightness of the image.

Three points will probably do it. If that is not accurate enough, add two more curve points, midway between the other points. I doubt you will need more than this, but you may of course continue adding curve control points until you no longer see any improvement in the match between the two
images.
This procedure is less satisfactory for color images, but you may try
doing
the procedure once for each channel.

Method 4
———-
Here’s a full blown procedure, that I think is overkill for a grayscale image, but it is very precise and will work well for color images:
1) create two or three eyedropper points on each image, in the critical areas that need to match. For example, quartertone, midtone, and three-quartertone. Sky is usually a problem for outdoor shots, so you may want to include two different points in the sky, say for cloud and clear sky.

2) write down the values for the image you are happy with, then use curves to make those points match in the other image.

Method 5
———-
Curvemeister offers a more efficient refinement of method 4, using a
process
called "Curve Extraction". You may create your eyedroppers in the first image, lock them to their colors, save them to a file, load them into your second image, and fine tune the eyedropper postiions on the second image. You may try this procedure out using only the demo, save the resulting
..acv
file, and apply the curve in Photoshop’s curves.


Mike Russell
www.curvemeister.com
www.geigy.2y.net

MacBook Pro 16” Mockups 🔥

– in 4 materials (clay versions included)

– 12 scenes

– 48 MacBook Pro 16″ mockups

– 6000 x 4500 px

Related Discussion Topics

Nice and short text about related topics in discussion sections