Monday, November 28, 2016

OCR and Scans

The Silver Vixen was out for the day with a friend. This left the Gorse Fox to his own devices (assuming Jasper the cat didn't intervene and demand attention).

It started with more scanning of photos and reviewing of videos and diaries to check dates and locations. This took a while but it did drag the timeline forward to mid 1990. This in turn raised the spectre of the missing diary entries for a a number of weeks. The reason that the diaries were virtually blank is that the Gorse Fox and the Silver Vixen were sending each other emails almost daily whilst the Gorse Fox was working in Poughkeepsie and in Austin. The Gorse Fox had the emails... but only as printouts.

The scanner burst into life and 64 pages of emails were consumed. Then the Gorse Fox started looking for an OCR product that would turn the images back into text. It took fair bit of research but in the end a MAC App was found and the 64 pages turned into pure text.

The rest of the afternoon was spent checking the OCR results and starting to ingest the emails into his private blog. There's a long way to go, but the Gorse Fox is very happy with progress. (In case you are interested, the OCR app is called LEADTOOLS OCR and seems to be very accurate so far).

No comments: