Any use OCR (Optical Character Recognition) software? - GoFuckYourself.com

rowan · 02-15-2007, 02:21 AM

I have about 200 pages of typewritten docs from the 1940s and 50s that I want to be able to computer index. Typing it in would take too long, and if I outsourced it then it would no doubt come back with mistakes due to the person transcribing it being unfamiliar with terms and certain words from that era. So, OCR is the way...

I tried something called SimpleOCR and it barfed completely, spitting out pure junk on a 90 word document... not a SINGLE correct word. Like, absolutely useless.

Then I tried Abbyy FineReader and it only made about 5 minor mistakes which were fixed in about 30 seconds. Definitely impressed!

What do you guys use?

RedShoe · 02-15-2007, 02:22 AM

What are the documents of?

rowan · 02-15-2007, 02:25 AM

Quote:

Originally Posted by RedShoe

What are the documents of?

Government correspondence,

*suspicious look*

... regarding a former army/navy explosives warehouse facility that I am researching. Having it text searchable is going to make it HEAPS easier to find things, at the moment I just have to shuffle through 200 jpegs to find what I need.

rowan · 02-15-2007, 02:35 AM

Left side original, right side detected & corrected

02-15-2007, 02:21 AM	#1
rowan Too lazy to set a custom title Join Date: Mar 2002 Location: Australia Posts: 17,393	Any use OCR (Optical Character Recognition) software? I have about 200 pages of typewritten docs from the 1940s and 50s that I want to be able to computer index. Typing it in would take too long, and if I outsourced it then it would no doubt come back with mistakes due to the person transcribing it being unfamiliar with terms and certain words from that era. So, OCR is the way... I tried something called SimpleOCR and it barfed completely, spitting out pure junk on a 90 word document... not a SINGLE correct word. Like, absolutely useless. Then I tried Abbyy FineReader and it only made about 5 minor mistakes which were fixed in about 30 seconds. Definitely impressed! What do you guys use?

02-15-2007, 02:22 AM	#2
RedShoe 赤い靴 call me 202-456-1111 Industry Role: Join Date: Feb 2001 Location: The Valley Posts: 14,831	What are the documents of? __________________ SPECIALTY COSTUMES • PROPS • FX Superheroes • Monsters • Robots PM for details For any manufacturing needs. Adult or otherwise. aka BonsHigh on Insta Bonsai weed plants

02-15-2007, 02:35 AM	#4
rowan Too lazy to set a custom title Join Date: Mar 2002 Location: Australia Posts: 17,393	Left side original, right side detected & corrected