Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 




      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[filmscanners] Re: Sharpening and JPEG/TIFF (was: Color spaces fordifferent purposes)

I think you should reread what Ken and I are saying - the effect of
sharpening is more visible in a low res image, no more no less. Your
lengthy explanation below is helpful in explaining why it is more visible.
Thank you for that. The original response sailed by his question and I was
re-emphasizing his observation.


At 05:23 PM 09/06/2002 -0700, you wrote:
>Don, your support of Ken is a bit misplaced. TIFF vs. JPEG is non
>sequitur, Anthony is correct.  This is about the pixels in the image,
>not about the file format in which it's saved.
>When an unsharp mask (a.k.a. sharpening) is applied to an image, it is
>enhancing the contrast of "edges" or areas of rapid level transition by
>lightening the light side, and darkening the dark side. This is done to
>the actual pixels in an image, independent of the image size resolution
>setting. The software analyzes the image and applies the sharpening
>effect within a certain pixel distance; this is the "Radius" setting in
>Photoshop's Unsharp mask, a typical distance could be e.g. 0.7 pixels.
>If you open your 27 MB file and the low res catalog scan, then apply the
>same unsharp mask to both, the edges enhancement is applied to the same
>distance in pixels around that edge in the image. These pixels, however,
>represent quite different distance in the image. If you view both at
>100% resolution, and both happened to have a narrow feature 3 pixels
>wide in both, they both would be appear "sharpened" exactly the same.
>Now, naturally, a feature 3 pixels wide in the low-res image would be
>something close to 15 pixels wide in the high-res image. Therefore the
>edge-enhancing effect would appear much more pronounced on the low-res
>one. If you compared them side-by-side, your catalog scan might be all
>visible in the window when viewed at 100%; while the high-res would have
>to be zoomed out to 20% of actual, and the sharpening effect would be
>miniaturized on screen and be far less noticable.
>This is why you should never apply the unsharp masking on your high-res
>scans until the final target use of the image is known, and, if
>necessary, the image is resampled down for that use.  For example, if
>you print a 360dpi image on a high quality inkjet printer on glossy
>media, you would need just a little unsharp masking, whereas printing
>the same image on offset press where the 4-color process screening will
>make images appear much softer you would need to apply a much stronger
>unsharp mask for the same final apparent crispness.
>If this same image was used for web, you would first downsample it to
>72dpi, then unsharp mask it for appropriate level of crispness at that
>Ken wrote:
>>>> ... but could someone offer a technical explanation
>>>> of why sharpening has so much more visible effect
>>>> on jpegs as opposed to TIFFs?
>At 10:22 AM 09/06/2002 +0200, Anthony wrote:
>>> It doesn't.
>On Sunday, June 9, 2002, at 07:46  AM, Don Marcotte wrote:
>> I support Ken. I'm currently scanning a large number of rolls of
>> negative
>> film. They are just 10x.6.67 inch by 72 ppi images for screen display.
>> I'm
>> keeping them in an electronic catalog of my images. Unless something has
>> changed in Photo Shop 7, which I recently acquired, sharpening is much
>> more
>> noticeable on these small JPEGs than on 27MB TIFFs that I use for
>> printing
>> or creating slides. I would like to emphasize the word "visible" in
>> Ken's
>> question.
>Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
>or 'unsubscribe filmscanners_digest' (as appropriate) in the message title
>or body

Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 


Copyright © Lexa Software, 1996-2009.