Perceived sharpness for sure. As to the number of pixels for a line I
think the answer is still two since we're only considering one dimension
(the thickness) of the line and not its length.
Chuck Norcutt
Doug wrote:
> On Sunday, October 21, 2007 21:53, Chuck Norcutt wrote:
>> Correct. I always forget about that line pair stuff. But I'd like to
>> see the actual comparison. Digital seems to surprise with images that
>> look like they have higher resolution than they should. As Moose's
>> tests with the 300D and 5D point out there's more to resolution than
>> pixels.
>>
>> Chuck Norcutt
>>
>> C.H.Ling wrote:
>>> You have to remember the resolution is lp/mm, you need at least two
>>> pixels per lp.
>>>
>>> C.H.Ling
>
> Wouldn't you have to have 4 pixels? It seems to me a line must consist of a
> black "line" next to a white "line". By resolution do you mean sharpness or
> perceived sharpness? I suspect that edge gradient has as much an influnence
> on percieved sharpness as absolute resolution. I remember working on a vision
> system at RPI in the 80's. We determined that you could "see" a filament that
> was about one fifth the size of the pixel size of the camera we were using.
> It darkened the pixel enough that in an uniform lighted field it was easy to
> threshold and do what we had to from there.
>
> ==============================================
> List usage info: http://www.zuikoholic.com
> List nannies: olympusadmin@xxxxxxxxxx
> ==============================================
>
>
>
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|