Below is what I saw in the thread http://groups.google.com/group/tesseract-ocr/browse_thread/thread/2321deb561450e76/554c7a8cec11c073#554c7a8cec11c073, where the last line mentions "nobatch". I am curious if you have any ideas about this. I recollect that in some case I got better results with nobatch and in some cases worse. There is a part in "Training Tesseract" that deals with handling characters that touch. It might have been relative to that. Does anyone know what the functional definitions of nobatch and batch.nochop are? Anyway, here is the quote from the thread: "From: unknowner <[email protected]> Date: Tue, 31 Jul 2007 05:53:23 -0700 Local: Tues, Jul 31 2007 5:53 am Subject: Windows Tool to edit boxfiles Hi guys, if your going to train an new language this tool might help you a bit. It shows the generated boxefiles over the picture. you need: - dot net 2.0 - picture as uncompressed tif - boxfile (tesseract.exe train.tif train nobatch makebox) in the same dir as the picture "
On Aug 24, 7:01 pm, 74yrs old <[email protected]> wrote: > re:"the thread on bbtesseract > talks about using "nobatch" in the command line" - where? > > > > On Tue, Aug 25, 2009 at 4:28 AM, SteveP <[email protected]> wrote: > > > Based on the tif you posted, it appears that the resolution is too low > > and the "*" characters are too small to give good results. The FAQ > > documentation suggests you double your resolution at least. That > > might be sufficient. > > > If not, relative to command line options, the thread on bbtesseract > > talks about using "nobatch" in the command line. I tried using > > "nobatch" instead of "batch.nochop" and it seems to make a > > difference. I tried both and compared. I did not really know how > > they work. Maybe someone else can comment ??? > > > The two "4"s in your image touch, so those need to be handled one way > > or another per "Training Tesseract". Ray Smith had a post in the last > > few days about using shorter substrings; that might help. > > > On Aug 20, 9:06 am, Chris <[email protected]> wrote: > > > Using Tesseract 2.04 through command line using the default eng > > > training files. > > > Attempting to make a box file off the following image. The first line > > > is read fine, but the second is ignored. > > > > Is there anything i can do, any parameters i can pass to adjust how > > > the box file is created? > > > > I have the tif posted at the link below, along with a zip of the tif. > > >http://omni19.tripod.com/recclaim/test.ziphttp://omni19.tripod.com/re...- > >Hide quoted text - > > - Show quoted text - --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
