@Quan Ok that with pam 0 OCR isn't performed but I'm expecting that when I run "tesseract input_image output_text -l eng -psm 0" I'll get the analysis response in the output_text file. With Tesseract 3.02 it isn't so :(
@zdenop So Tesseract v. 3.02 doesn't support this feature... I'll try 3.03 version! Many thanks! Il giorno domenica 11 maggio 2014 13:53:45 UTC+2, Quan Nguyen ha scritto: > > With psm 0, Tesseract does not perform normal OCR function but analyzes > layout; it produces such characteristics as Orientation, Writing Direction, > and Textline Order. Check Tess4J unit tests for usage of OSD. > > On Sunday, May 11, 2014 5:48:39 AM UTC-5, Joe Aspara wrote: >> >> I'm struggling with the OSD function of Tesseract 3.02. >> I tried the standalone version via command line and the Tess4J version >> too, but I always obtain an error with different input types. >> >> I downloaded the osd.traineddata for version 3.01 (I guess no such file >> still exist for v3.02) from here >> https://code.google.com/p/tesseract-ocr/downloads/detail?name=tesseract-ocr-3.01.osd.tar.gz&can=2&q= >> and I copied it properly in the TESSDATA folder >> >> Below my experiments: >> >> COMMAND LINE >> tesseract input_image output_text -l eng -psm 0 >> response: Error during processing. >> >> With psm = 1 it read text with very bad quality, with psm = 2 or 3 it >> give my empty output. >> >> As far as I know only 0 and 1 values perform OSD! From the reference: >> 0 = Orientation and script detection (OSD) only. >> 1 = Automatic page segmentation with OSD. >> >> >> TESS4J >> Tesseract instance = Tesseract.getInstance(); >> instance.setLanguage("ita"); >> instance.setPageSegMode(TessPageSegMode.PSM_AUTO_OSD); >> String result = instance.doOCR(myImage); >> >> result always is empty at the end >> >> To know the input orientation it's critical for my project but at now I'm >> not able to find a way to accomplish this. >> >> I hope somebody can help me! Thanks in advance >> >> -- 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]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/f8f0e275-9c21-4a8e-80a4-06ecd0760bcd%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
