| [Thread Prev] | [Thread Next] | [Thread Index] | [Date Prev] | [Date Next] | [Date Index] |
Lots of people seem to be doing this sort of thing with their images
but I know of no published reference on it yet--at least in the
paleo. literature. It's very easy. Basically you need a series of
images of an object that have been taken at different focal depths.
Once you have those you start with the image shot at the lowest focal
depth, encircle the areas that are 'in focus' on the next image in
the series, and then drag the selected areas over the basal image,
adjusting the position of the slice as needed. work your way up the
series in the same manner, If your original image set is good
('in-focus' areas overlap) and your eye serves you well you should
end up with an image that shows a greater focal depth than any of the
original images. Of course, you can achive the same result (to some
extent) with high-end lenses. But this method is much less expensive.
There are bits of software and hardware available for automating the
process, but you can get good results without resorting to these. As
you suspect, Photoshop is the workhorse for this process, but a
variety of software can be used to do the same thing.
I hope this helps.
Norm MacLeod
>Dr. MacLeod, I was cleaning out my mailbox when I came across this
>old email. I saved it because I was very interested in how you
>combine the images by "hand" to increase the depth of field. I
>assume that this is done in Photoshop or a similar graphics package.
>Is there a source you can point me to that explains how to do this?
>Thanks David Waugh
>
>>A very interesting string about digital images. Just a short note
>>here about the image compositing process that was discussed a couple
>>of days ago. We've used image compositing to correct depth-of-field
>>problems on both macroscopic and microscopic images for years, most
>>notably in the image sets created for the PaleoBase databases. We're
>>very pleased with the results, examples of which can be seen a the
>>PaleoBase web site (http://www.paleobase.com/home.html). NHM authors
>>(and others who have asked us to create these images for them) have
>>also published many such images in the technical literature. They are
>>absurdly simple to create and do not require expensive technology.
>>Several years ago I conducted a test between our hand-made images and
>>the Synoptics image compositing software for some moderately
>>difficult microscopic images (planktonic foraminifera). The hand-made
>>images were quicker to produce and achieved a better result than the
>>software (a least at that time). I'm seeing automated image
>>compositing routines appearing in lots of image analysis packages
>>just now so it looks like this is going to be a standard
>>image-manipulation feature in the near future. Still, I think better
>>results can be achieved by an analyst who has a 'feeling' for the
>>geometry as opposed to an algorithm that is maximising/minimizing
>>some number. I suppose it also depends on how much time a person
>>wants to spend at this sort of work.
>>
>>The PaleoBase databases (co-published by Blackwells and the NHM) will
>>be 'on show' at the GSA meeting next week. Anyone interested in these
>>composites can see them there.
>>
>>Norm MacLeod
>>
>>
>>--
>>
>>
>>___________________________________________________________________
>>
>>Dr. Norman MacLeod
>>Keeper of Palaeontology
>>The Natural History Museum,
>>Cromwell Road, London, SW7 5BD
>>
>>(0)20-7942-5295 (Office)
>>(0)20-7942-5546 (Fax)
>>
>>Web Page: http://www.nhm.ac.uk/palaeontology/a&ss/nm/nm.html
>>
>>___________________________________________________________________
>
>
>--
>-------------------------------------
>David Waugh
>Department of Geology
>Kent State University
>Kent, OH 44242
--
___________________________________________________________________
Dr. Norman MacLeod
Keeper of Palaeontology
The Natural History Museum,
Cromwell Road, London, SW7 5BD
(0)20-7942-5204 (Office)
(0)20-7942-5546 (Fax)
Web Page: http://www.nhm.ac.uk/palaeontology/a&ss/nm/nm.html
___________________________________________________________________
Partial index: