Views
161
Replies
3
Status
Closed
One very common question I get asked is how to turn a halftone image into a continuous tone image (or at least simulate one). There are some techniques that work better than others, usually dependent on the subject matter and the type of halftone, but I had an idea using a new PScs feature that might work for all halftones, if I can just figure out how to implement it.
Halftones are, by definition, a grid with a dot in the middle. The darker the value of that particular grid square, the larger the dot. I was thinking that if you could divide an image up into its constituent grid (either literally or virtually) and then run the new Blur>Average filter on each grid square, you’d end up with the closest-possible approximation. Essentially each grid square would end up being a virtual pixel (though each virtual pixel might actually be made up of several identical real pixels).
Obviously this would take some planning, not to mention some computational power. But I do think it would be worthwhile, especially if there was some sort of GUI way of setting it up (rather than simply by math or trial and error). I wonder if this would be possible with an action, or would it require a script? Or even further, would we be in filter programming territory?
Right now I can’t even envision a reliable way to do this manually, so I’m throwing it out for discussion. Given a random halftone image, with no known screening value, how would you go about determining the grid placement, selecting one grid square, moving to the next, figuring out when you’re at the end of a row and moving to the next row, etc.?
—
– Doug Nelson
==============================
http://www.retouchpro.com — the #1 online community for retouchers and restorers
Halftones are, by definition, a grid with a dot in the middle. The darker the value of that particular grid square, the larger the dot. I was thinking that if you could divide an image up into its constituent grid (either literally or virtually) and then run the new Blur>Average filter on each grid square, you’d end up with the closest-possible approximation. Essentially each grid square would end up being a virtual pixel (though each virtual pixel might actually be made up of several identical real pixels).
Obviously this would take some planning, not to mention some computational power. But I do think it would be worthwhile, especially if there was some sort of GUI way of setting it up (rather than simply by math or trial and error). I wonder if this would be possible with an action, or would it require a script? Or even further, would we be in filter programming territory?
Right now I can’t even envision a reliable way to do this manually, so I’m throwing it out for discussion. Given a random halftone image, with no known screening value, how would you go about determining the grid placement, selecting one grid square, moving to the next, figuring out when you’re at the end of a row and moving to the next row, etc.?
—
– Doug Nelson
==============================
http://www.retouchpro.com — the #1 online community for retouchers and restorers
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.
