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

Re: Image Processing Problem (a reply)



To:	Terry Arnold



Terry,

You recently wrote "A colleague of mine is in the hroes of fininshing a 
monograph on the radulaof Cyraeidae. This monograph is illustrated with several
hundred SEM and optical microphotographs. ....... "

I don't know if the following is likely to be of any use to you or your 
colleague but here goes:

I'm a Brit avocational paleontologist working with a range of fossil groups
from the lower Cretaceous (Albian) Gault Clay of South-East England.

1.	Working with a CCD Camera connected to a PC

Mainly (but not exclusively) for scanning tiny shark teeth (from 0.5mm upwards)
I have recently set up a 500,000 pixel CCD camera onto a Meiji trinocular
low-power microscope illuminated by a Schott fibre-optic KL1500.    This is
connected via a Video Logic frame grabber (Captivator Pro for VMC) into my PC
(a custom built Pentium 133 with a big hard drive (4Gb)) using a software
programme called "TV-Snap" which is, in turn, connected to a Lexmark Optra R
plus 1200 DPI grey scales laser printer.

The image from the CCD onto the screen is excellent.   Everything depends on
how you set the lighting up.  I reckon they are comparable with those produced
with a regular optical camera (other than the highest quality cameras) and
exceptionally with an SEM.

The TVSnap programme lets you "snap" the image, then select an area immediately
around the fossil, snap the area then save the snapped image to a file. 
Because the images are individually quite small (0.5Mb to 0.8Mb) I move them
via the clipboard into a suitable programme (A Corel programme, Paint Shop Pro
or into a Word 7 document).  I tend to use the Word 7 option most because it is
clean and convenient - by setting up a page format then using the "Insert
Frame" option and pasting into the frame.  You can size it, move it and so on
and by double-clicking the image can rotate, insert text etc.  If I need to
edit the image then I use the tools in Corel Draw, Corel Photo Paint or PSP.

With practice and a few good tips I'm able to produce some near photo-quality,
images of the teeth (from between 0.5mm up to 1.5cm) and the depth of field is
also pretty good although I have still some way to go.

I print the images directly onto quality paper such as used in copy shops on
colour photocopiers (in the uk costs around pounds sterling twenty-three for
five hundred sheets).   Paper with a shine to it gives an even better
photo-like image.

The Optra R plus is a true 1200 dpi printer.  I've been using it for about four
months now and have been surprised with the quality of the output (as has
anyone who has seen it).  It takes around two minutes to print out a 6Mb A4
page (which would usually have around eight views on it).

Incidentally, I also get very good results with other tiny fossils that I am
recording (gastropods, ammonite protoconches, fish otoliths you name it and
even individual forams and ostracods come out well (the latter nothing like
true SEMs) but the best results tend to be around the 4 to 10 mm size.

2.	Working with a HP Scanjet 4C 2400dpi flatbed scanner

For larger fossils (from about 2cm upwards) I use a flatbed scanner.  It lends
itself to practically any size that will fit onto the scanner and works with
all fossil groups from teeth, gastropods, lamellibranchs and so on.  The only
limitation is the power of your PC (Speed, Ram and Memory). Here its best to
use an example so if you will forgive me:

I'm very interested in lower Cretaceous (gault) ammonites and have been for as
long as I can remember so I am spending a lot of time scanning and producing
plates of specimens from my and some friend's collections.  These tend to be
quite small (between 2cm and 7cm - the straight heteromorph ammonites up to
12cm).  The ammonites vary from unornamented planulate to very inflated 
tuberculate - the heteromorphs exhibiting varying degrees of coiling and 
curvature.

The HP 4C working through the PC and out again via the Lexmark is really quite
remarkable!  It copes with lateral and dorsal views of the ammonites in much
the same way as an optical camera.  The depth of field on the side of e.g. an
inflated species of Euhoplites (say, Euhoplites proboscideus, Spath) is every
bit as good as any photographic image when scanned, zoomed, scanned again and
displayed on the screen particularly when viewed in millions of colours.

The rub is that the images can take individually a lot of memory.  Scanning a
5cm dia ammonite at 1200dpi as a sharp black & white photo usually takes around
5 to 6Mb so a plate of four ammonites is up to 24Mb and this begins to take a
lot of time to move around and format for the printer.  Having said this a 5cm
ammonite scanned at 1200dpi when put into Corel Photo Paint would fill half
your room so the difference in print resolution between a 5cm ammonite scanned
at 600 and 1200 is barely discernable and 300 dpi produces a pretty good image
too.

To create a more permanent personal record I am having the plates of ammonites
put onto CD ROM.

3.	Your friend's monograph

The flatbed scanner procedure should work well with his photographs and if tied
in to a 1200 dpi printer I guess would do the trick.  I will experiment with a
few of my own Scanning Electron Micrographs these next few days to see how it
works.

Finally, I'm still very new at this and am learning all the time.  If you would
like me to send you (or any other interested party) a paper copy of the kinds
of images I'm obtaining then I'd be happy to oblige - just let me have your
address.

Regards

Jim Craig
"Toliapicus"
Augustine Road
Minster-on-Sea
Isle of Sheppey
Kent ME12 2LZ
UK

e-mail:	jim.craig@dial.pipex.com
	(hv89@dial.pipex.com)
tel:	+ 44 (0) 1795 875542