New Mitsubishi sharpening technology

padupadu Registered Users Posts: 191 Major grins
edited April 19, 2007 in Finishing School
http://padu.merlotti.com
http://padu.smugmug.com
www.merlotti.com
Sony dslr A100, Minolta Maxxum 7000, Voighlander Bessa R and Calumet 4x5 View Camera

Comments

  • LiquidAirLiquidAir Registered Users Posts: 1,751 Major grins
    edited April 19, 2007
    I have read the papers on the plenoptic camera developed at Stanford. That group was recording something akin to a hologram which they could focus at to distance in software.

    It looks like what this group is doing is replacing the standard aperture with a special mask which probably looks something like a grid of pinholes. Remember that out of focus points of light take on the shape of your aperture and that the blur of an out of focus image is created by overlapping many images of your aperture (that is why depth of field increases as the aperture gets smaller). With this special aperture out of focus pattern is a grid of small dots rather than a single larger circle and the spacing of the grid points indicates how far out of focus the point is. My guess is that this special grid pattern lets them recognize out of focus hard edges because the overlapping dot patterns leave behind a recognizable signature. The process is likely similar to how modern auto-focus systems determine the proper focus distance. The price of this refocusing process is likely to be significant increase in noise.

    Note: from a mathematical point of view this description is not terribly accurate, but it is the best I can do without resorting to a bunch of jargon from signal processing theory.
Sign In or Register to comment.