Blue light used for increased contrast and detail. The Canon T3i is a poor choice for this type of lighting but did better than I expected. ISO 400-800 used to make video. Its much more noisy than I like. I'll try this slide again with UV and a monochrome camera... The stitching is pretty bad. For some reason MS ICE wasn't able to stitch the two photos.
I have also experienced difficulty in persuading Microsoft ICE to composite seemingly straightforward diatom stitches. Where ICE does work the 'joins' can be uncomfortably obvious. The highly detailed and strongly repetitive structure with relative lack of distinctive features may be a factor. Small differences in focus point between adjacent images can also throw ICE off task. I have found that very generous overlaps between individual images to be stitched helps to overcome the problem. Attention is also needed to ensure all images show as even a focal plane as possible given the constraints of high power shallow focal depth, and the potential additional complicating factor of stitching stacking outputs.
Last night I tried again 455nm light and a cheap monochrome security ccd camera with much better results. I also tried 365nm light. Bright field worked well but there was not enough light for DIC. I'm going to attempt to mount a reflector and possibly a different collimator lens to focus more light. I'll try to post some results tonight.
Dave, this is the first time I've tried a diatom stitch with ICE. ICE has worked great for other microscopic subjects for me. I thought the diatom would be an easy stitch.
Pau, I'll probably going to wear the coverslip down on this slide before I'm done with it.