35mm microfilm test for the Toronto Camera Club

 

SELECT ORIGINAL UNDER THE IMAGE TO SEE FULL RESOLUTION

IN EACH IMAGE PAIR TH SECOND HAS BEEN ENHANCE FOR TEXT

 

 

Images starting with 35mm _silverxxxx are from a different source and I added these to see the results from an old newspaper page (not too promising) and some 1872 type-written pages.

 

Notes on TCC film

Sprocketless, B&W negative

Scanned at 5.2Mp with film scanning machine

Manually held in place with modified slide adaptor (no, the film was not cut).

Framing was not perfect and large white bar on the left was actually edge of device I used to hole the frame.

Images were processed with 12/16 bit resolution and then published -> BMP -> jpeg.

I did not for a heavy white black saturation as I found them easier to read in this form.

Most,. but not all of the frames had two pages per fame. You can see and example in this set. Sometime notes were laid across the pages.

 

The entire frame was initially stored as a raw pattern

All of the images here were processed with a recipe, no individual adjustments were performed.

 

Each image appears twice, the second having some enhancement to bring our edges of the text.

Generating multiple versions of each photograph is not a problem.

 

Improvements that I can make

·                   Eliminating debayering in processing, this should in theory, slightly increase resolution.

·                   Feed the film with movie film rollers as opposed to test with slide adapter – to ensure optimum focus

·                   Framing could be improved to eliminate the white bar on the right, and focal distance changed to cover largest possible page height.

 

If you wanted your micro film digitized with this system then:

 

·                   I would recommend capturing each side of the frame separately, for resolution consideration, and possibly the center of the fame.(which I would not count as a separate page).

 

I have not examined all of there film however.

·                   It is not convenient to switch focal length on a frame by frame basis. If I set up to capture a page per side or center, and image with fills the whote frame, would be captured in overlapping seconds.

·                   The identification banner at the bottom of the frame appears to be there most of the time, and I would use this for indexing. If it is absent I will have to manually move to the next frame.

·                   I noticed all is not black and white in the microfilm world. I like the images better as grayscale, boosting the contrast currently may result in messy parts as the black level varies across the negative.

·                   Yes there and pens holding down pages.

 

Please let me know what you think of these scans and if you want to see any image processing changes tried on the test images.

john@bitworks.org