Ron Parker wrote:

> I find it interesting that every image processing program I've ever seen
> has a filter to differentiate the image along a direction (usually called
> "emboss") but none seem to be able to do the inverse operation; that is,
> integrate the image along a direction.  If you could, you could then take
> a carefully-controlled photograph of a penny (might need to paint it flat
> white first, to remove the specular highlights) and use the integrate
> feature to recover the contours.

    Or just put a penny on a scanner.

    As to and inverse operation, I don't see what you mean. But there is
always a separate inverse function, the photonegative.

    And there is also almost always the "stamp" filter instead of emboss
which often works much better for things where emboss appears to be the right
thing but doesn't work. And with stamp you should get control of edge depth
and direction.

Post a reply to this message