It looks like your hot spots are a square of 4 pixels, surrounded by a 1 to 2 pixel frame of "weirdness". What's wrong with creating a mask that is 100% for the center square, and some lesser percentage for the outer frame and spotting it with an action?
We're looking at a very small part of the picture, and traditional dust N scratch techniques (except for hand spotting with the clone tool) pretty much make this stuff disappear. I'd sacrifice spotting "perfection" for the speed of an action. I guess you wouldn't.
I've fooled around a lot with spotting, as my scanner doesn't have ice. The techniques I use to combat dust and scratches do not depend on color or intensity of the "warts" as long as the location of the warts is defined in a mask. Even if at times a given hotspot is unobjectionable, why not batch process it anyway?
Am I still missing something here? Or am I less of a perfectionist?
Rob Studdert wrote:
On 7 Oct 2003 at 9:29, Lon Williamson wrote:
If the location of the pixels is constant, the color they take on is a non-issue, even if the color varies. You could develop a mask specific to your sensor and whack them out using a median filter of some suitable radius. And you could set it up as a batch action. Much faster than manual touch-up.
If it were that simple I would have done it. Intensity varies with exposure plus the hot pixels also develop a colour fringe which changes colour based on the colour of the adjacent pixels. This means that unless carefully spotted they will be visible, especially on a 4MP sensor like mine.
See the following crop from a EI 320 1/2 second exposure (EXIF info intact):
http://members.ozemail.com.au/~geroc/P9095721.jpg
Not a perfect example as it's none too sharp however you get the idea.
Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
