Microscopy With A Film Scanner
NickFitz writes "If you've got a film scanner lying around, this site shows you how to use it as a microscope. "Your monitor displays images at about 70-90 dpi, so taking the example of my Mac monitor at about 75 dpi, we get a magnification of 4000/75 which is about x53. It's not a lot and isn't going to show the likes of blood cells, but it should give an interesting view of small transparent objects." Did you know that bees wings are hairy?"
Want photographs? You can get surprisingly good results by simply holding a digital camera flush to the eyepiece. I have a few of these I did for fun here.
Have fun.
Roving Web-Teleoperated Robot
People have been using scanners in this way for image forensics for years. We also used this method to get around image mosaicing large microscopy fields several years ago in an undergraduate biology class I took. (amazing what students come up with to avoid work) Now of course image mosaicing software is available to get around this problem, but it's good to see science stuff like this get out to the main stream. Perhaps this will also make it into a few junior high and high school classes as a cool exercise.
Visit Jonesblog and say hello.
Since this seems like the type of site that will be slashdotted, I cached a copy of the bee wing image. Enjoy!
Why do I h8 apple?
.. has also been done, albeit with a lesser degree of success. I believe it appears in the Annals of Improbably Research collected book.
Or buy the Intel computer micriscope for 1/10th the price of a really bad shape used film scanner and get better results.
This is great if you have the hyper-expensive device...otherwise buy the intel microscope and call it done with better images.
These things cost like $500 new at the cheapest. I think that even used you would still be better off just buying a cheap microscope that gets 400x like this one for $134.
Building a Digicam from Scanner Elements links to this page
Image quality might suck because most scanners have lighting systems that are appreciably brighter than the image projected by a lens. So the image might be dark. Moreover, the coupling of the imaging lens' projected image into the scanner optics would be imperfect and so the center of the image would probably be far brighter than the edges. Finally, the hacked scanner-camera would only work on motionless objects -- any movement during the long scanning time would create interesting artifacts.
But it would be quite cool because you would have an umpteen megapixel camera. A 300 dpi flatbed would create an 8.4 megapixel image and a 600 dpi flatbed would provide a 33.6 megapixel image. A 4000 dpi slide scanner would lead to about a 21 megapixel camera.
Two wrongs don't make a right, but three lefts do.