On Friday, April 4, 2014 7:36:09 PM UTC-4, Quan Nguyen wrote: > > Make sure your images are of at least 300 DPI. > > In a convoluted fashion, I printed out the image and then scanned it into my scanner which allows me to set the dpi on the image, which I set to 600dpi. The image was a bit skewed but I decided to give it a try anyway. I did not believe that there was any change this would work, as it was essentially the same image. Well, I was wrong. The translation came back pretty good, a few errors but WAY better than the original translation.
Doing some additional googling on how to increase dpi without going the whole print & rescan route, I came across ImageMagik and the 'convert' command line tool. I ran the following on my original image: convert -units PixelsPerInch Screenshot.png -resample 300 Screenshot300.png and then tesseract Screenshot300.png out300 It came back with a translation that was pretty darned good- only one mistake. I bumped it up to 600 but got the same result. I am very pleased with the result. I may try upgrading again to the latest version now that I know how to get the most out of this and see if it fixes this one mistake. Thank you for your suggestion! I had read that suggestion in the docs prior to my attempts but didn't believe that taking the same image and simply bumping up the dpi on it would fix the translation errors. It seems like this would be something that tesseract would do on it's own in the image pre-processing phase. -- -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en --- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
