DPRG
DPRG List  



[DPRG] Camera needed at RBNO

Subject: [DPRG] Camera needed at RBNO
From: Earl Bollinger earlwbollinger at comcast.net
Date: Mon Oct 20 21:04:01 CDT 2003

I don't think anyone in the club can afford a 7.2megapixel camera for
that resolution. My pitiful Sony camera only goes up to 1600 x 1280.
You could take 35mm film shots (or use that big portrait camera), and
them scan the photos in a 300dpi on a good scanner. You could even geta
few 8x10 blow ups and scan those in at 300dpi.


-----Original Message-----
>From: dprglist-admin at dprg.org [mailto:dprglist-admin at dprg.org] On Behalf
Of William James
Sent: Monday, October 20, 2003 5:19 PM
To: DPRG
Subject: Re: [DPRG] Camera needed at RBNO

Thanks, to everyone that replied to my email. I guess
if it specified that they want it in 300dpi, but don't
specify a size. So if I could get someone that has a
good digital camera to come to the RBNO and take a
pic, I would appreciate it. That way I can get a good
sized 300 dpi photo.

Bill

--- Chuck McManis <cmcmanis at mcmanis.com> wrote:
> At 12:49 PM 10/20/03 -0700, William James wrote:
> >I need to take a 300dpi picture of Beerbot. My
> phone
> >camera has a resolution of 640 x 480. I don't
> believe
> >that is enought. [sic] First, how do you figure out
> >the DPI of a specific resolution.
> 
> This is a very common area of confusion ...
> 
> Cameras don't have a "dots per inch" resolution,
> they have a resolving 
> resolution. Printers on the other hand have a
> "printing" resolution.
> 
> Imagine a camera, the focal plane of the imaging
> element has some number of 
> pixels (in your case 640 x 480). If the width and
> height ratios of those 
> pixels were exact (they aren't but assume for the
> moment they are), and the 
> camera was focused on a robot 18" away, such that
> the robot, whose 
> dimensions were 12" on a side, filled the imaging
> area. (slightly less than 
> a 90 degree field of view) Then with your camera in
> that configuration you 
> can "resolve" things that are as small as .02" on a
> side (that's a little 
> less than 1/32nd of an inch).
> 
> As it is unlikely that your robot has many features
> this small, imaging it 
> at 640 x 480 would provide acceptable resolving
> resolution for most 
> applications.
> 
> Then there is the issue of reproducing this image in
> print.
> 
> Printers, and especially color printers, have a
> "printing resolution" 
> (usually specified in dots per inch) that identify
> how accurately those 
> printers can place their ink. For example a 300 dpi
> printer, can print 
> black and white ink dots that are slightly more than
> 3 mils apart. 
> Technically that makes it a 150 lines/inch printer.
> 
> Now when you get to color, these printers often use
> a technique where they 
> overprint dots in a several base colors to create a
> new color. The 
> interesting (and glossed over) fact is that the more
> "room" they have to 
> print their inks, the more colors they can
> reproduce. The resolution at 
> which they can print their full gamut is called the
> "screen resolution." 
> Gamut is a term for the set of all possible colors
> that can be reproduced, 
> and screen resolution comes from the days of
> silkscreen printing where the 
> holes in the screen determined the size of the
> printed dots. Interestingly, 
> an HP 970 which is called a 300DPI printer has a
> screen resolution of about 
> 125 DPI. You can demonstrate this to your self by
> using a program like 
> Corel Photopaint (or Adobe Photoshop) and create two
> versions of an image, 
> one at 300 DPI and one at 125 DPI. Print them both
> to your printer, they 
> should be exactly the same size on the page, with
> one having better color 
> fidelity than the other. If you've got a standard
> color wheel type image 
> its best, but anything with a lot of color variation
> will do.
> 
> Now getting down to brass tacks...
> 
> If you have an image that is nominally a 72 DPI
> image (CRT screen 
> resolution for a 640 x 480 screen, 13" diagonal
> display area) You can 
> re-sample it in most imaging tools to change its DPI
> to something you would 
> find more appropriate for printing (100 - 200 DPI).
> Different programs do 
> this in different ways, however remember that for
> printing you want both a 
> "size" (which is under your control) and a "dpi"
> which is tuned for the 
> printer you are printing to. The number of pixels
> will be "dpi * size in 
> inches" and if the number of pixels in the resampled
> image is larger than 
> the original, artifacts in the form of "blockyness"
> can be introduced. When 
> the resampled image has fewer pixels than the
> original you can get 
> artifacts in the form of "aliasing."
> 
> So if the consumer of your image wanted to print it
> as a 2" x 1.5" image 
> next to a paragraph describing your robot, then
> resampling a 640 x 480 
> pixel image to an image that is 2" x 1.5" and prints
> at 300DPI (600 x 450 
> pixels) will be able to use your image with no
> change. If they want to 
> print it at 4 x 3" (1200 x 900 pixels) then they
> will have to resample it 
> to the larger size. Depending on their resampling
> algorithm this will 
> either make it look pixellated or fuzzy.
> 
> Another consideration is that typically you will
> crop the image before you 
> print it, and cropping takes away pixel information.
> Typically a raw image 
> size of 2048 x 1600 is usable in nearly any media
> short of film projected 
> in a theater environment.
> 
> I know, a lot more than you asked for and probably
> not very helpful, I'll 
> stop typing now ...
> 
> --Chuck
> 
> _______________________________________________
> DPRGlist mailing list
> DPRGlist at dprg.org
> http://nimon.ncc.com/mailman/listinfo/dprglist


__________________________________
Do you Yahoo!?
The New Yahoo! Shopping - with improved product search
http://shopping.yahoo.com
_______________________________________________
DPRGlist mailing list
DPRGlist at dprg.org
http://nimon.ncc.com/mailman/listinfo/dprglist



More information about the DPRG mailing list