PuppyOCR now with GUI interface

[tronkel]

Here is another better designed update for PuppyOCR
Size: 2.1MB+

Get it here:
http://www.datafilehost.com/download-00da7027.html

There is now no need to start XSANE externally.
If you have a scanner available to the system, simply click the red SCAN button.

Your document will then be scanned in and OCR'd with one click. SCANIMAGE now does its stuff in the system in the background. The scan automatically produces an image called a.tif in /root. The OCR stage will then create a text file from this called a.txt - also in /root.

No need now to mess with converting files produced by XSANE scans

Click VIEW and a GEANY window pops up and displays your OCR'd text.

If you need to directly access an image file because say, there is no scanner access available, you can still OCR an image file by entering the paths to the input and output files and then clicking on the OCR IT! button.

I daresay other stuff will occur to me as well. so expect further updates

[gcmartin]

... need to directly access an image file because say, there is no scanner access available, you can still OCR an image file by entering the paths to the input and output files and then clicking on the OCR IT! button. ...

@Tronkel, this is of interest.

Is there a list somewhere of which image filestypes that it will convert to text?

Is this the same

Thanks for this good work.

[tronkel]

gcmartin wrote:

Is there a list somewhere of which image filestypes that it will convert to text?

Since Tesseract is the OCRing engine for this program, I have hard-coded the system to handle only 'tif' files for 2 reasons.
1. The Tesseract engine can handle both PNM and TIFF file types but seems happier with the TIFF format.
2. If you need to scan for a PNM, you can use XSANE which does a good job with the PNM types, but appears to have some problem with saving to TIFF. Maybe XSANE should really be recompiled with the TIFF option enabled.

Between PuppyOCR and Xsane you're covered for both at the moment. I'm not sure if the XSANE project is actively being maintained or not, but it functions mostly OK running in Puppy.

The other alternatives that your link points to look interesting. I'm not familiar with them as yet, but will have a look at them when I get the time. Maybe they have some good features that I could include in PuppyOCR.

There will most likely be an on-going update cycle for PuppyOCR as new ideas occur to me. Any feedback and ideas welcome.

[disciple]

You do know that Tesseract can support all kinds of image file types now, don't you?