Pixel density revisited

Started Oct 22, 2008 | Discussions thread
ejmartin Veteran Member • Posts: 6,274
Re: There is a resized comparison

bobn2 wrote:

Missed that. Thanks. Unfortunately, its usefulness is compromised by
use of Photoshop resize, which I'm told (by John Sheehy amongst
others) decimates rather than resamples. The result is you don't lose
the noise but you do lose the detail.

Indeed. I just did the exercise I linked to earlier in the thread -- opened a new canvas in Photoshop, filled it with middle grey, added gaussian noise; then did a bicubic resize to half the number of pixels (downsize by linear factor sqrt[2] 1.4) in photoshop. The per pixel noise actually went up in the resized image.

I believe the problem is noise in the original which is beyond the Nyquist frequency of the downsized image; it seems that photoshop to some extent compresses that noise power into the spatial frequency spectrum of the downsized image. A small radius Gaussian blur (std dev roughly equal to resize factor) performed prior to the resize dumped the high frequency noise power, so that the resized image had reduced noise instead of increased, without any additional loss of resolution.

-- hide signature --
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow