OCR Software -- what's the best these days?

[amandagmu]

What's the best out there these days for Chinese characters? Also, I have a bunch of PDFs I already scanned in... do I need to convert them to images to run through OCR?

Basically, I want readable text in the end (even if I have to check things), goal being I can quickly look up characters by copying and pasting into online dictionaries and translators. ReadIris 12 for Asian languages looks pretty nifty but it's $199 without a decent demo. Anyone have an opinion?

Thanks!

Amanda

[gato]

Haven't used OCR for a long time, but did try 紫光 OCR a few years and it worked pretty well. The thread below says that Hanwang's OCR has the best recognition rate, which seems credible since Hanwang makes the leading handwriting recognition software (it's the engine used in Pleco).

http://www.51nb.com/forum/viewthread.php?tid=641158

请教中文的OCR软件哪个最好？

http://dl.pconline.com.cn/download/54490.html

汉王 PDF OCR 8.1 简体中文版

[chalimac]

I have checked ReadIris 11 with asian add-on and it works well. But with small characters there are always mistakes.

[PhilipLean]

I recently tried a demo of Finereader for Chinese - the result was not good.

ReadIris do not release a demo of their Asian languages ocr but you can contact one of the reseller shops and send them some samples to OCR, the turn around is about 24 hours. http://www.scanstore.com/Users/New_User.asp?LiveSupport=ON

they advise - "Please scan at 300 dpi, in Black & White and to TIF or PDF"

ReadIris did a reasonable job (30% accuracy) on some very difficult - poorly scanned text - I am looking for a better scanned copy of that text before I try again.

It should do much better on good quality scans.

Send them your samples and see what they can do. Please let us know. I have not bought an ORC program yet but I plan to in a few months, so I am still looking.

Edited September 28, 2009 at 03:55 AM by PhilipLean
added more detail

[imron]

If you're living in China the other alternative is find a student and pay them 10 rmb an hour to type everything up.

[amandagmu]

Thanks everyone! Not living in China... rates in Northern California are slightly higher, ha. Anyways, will let you know what works best soon, I hope!

[ABCinChina]

I had readiris and it did not work that great at all for me. I've personally given up on Chinese OCR.