[tesseract-ocr] Re: Systematic way to program and train tesseract OCR with artificial neural network and c++

2015-04-21 Thread Lavar Askew
Hi Harsha, I believe that Tesseract already implements a neural network. In fact it uses a convolutional neural net. If you download the source and look in the tesseract/neural_networks/runtime and tesseract/cube directories hopefully that can get you started until someone with a more insight

[tesseract-ocr] run-time error about missing functions, etc.

2015-04-21 Thread SDSL
Hi, I need to build tesseract-ocr (https://code.google.com/p/tesseract-ocr/) from source in order to OCR some PDF files. Many people use "convert" of imagemagick to first convert a PDF to a TIFF then resort to Tesseract to OCR the TIFF to a text file. Since Tesseract depends on the Leptonica I

[tesseract-ocr] OCR From Small Graphs

2015-04-21 Thread Aaron G
Hello - First, thank you to everyone supporting this tool... I've had pretty good success with it in the past, but am running into an issue, I'm hoping someone may be able to help with. I receive images similar to the attached, and based on what I have read, it sounds like tesseract may have

[tesseract-ocr] Tesseract training for korean country

2015-04-21 Thread chan
Hi everyone i'm trying to train for korean character by following the procedure from this link https://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3 but i'm stuck, could any one guide me to train for korean character .. -- You received this message because you are subscribed to the

[tesseract-ocr] Re: Multiple tifs to one file

2015-04-21 Thread Quan Nguyen
Try *Merge TIFF* function of VietOCR . On Tuesday, April 21, 2015 at 4:58:19 AM UTC-5, Stathis L. wrote: > > Is there a way to combine several tiffs into one ocr file? If so, what may > be the command for that? > > Thanks > -- You received this message because you are sub

[tesseract-ocr] Re: Processing an image

2015-04-21 Thread Craig Fairhurst
-psm 7 seems to do the trick but no legible data is returned On Monday, April 20, 2015 at 4:49:29 PM UTC+1, Craig Fairhurst wrote: > > Hi I want to prove its possible to break an image captcha verification > field with tesseract-orc, can you help? > > The image is > http://arbiter.rogues-allianc

[tesseract-ocr] Multiple tifs to one file

2015-04-21 Thread Stathis L.
Is there a way to combine several tiffs into one ocr file? If so, what may be the command for that? Thanks -- 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 tes

Re: [tesseract-ocr] Way to set minimum font size to reduce errors

2015-04-21 Thread Dmitri Silaev
You can try "textord_min_xheight" (in pixels). However it might not work as there are tons of possible reasons why that happens. There can be other params that could help. Show a few sample images and explain what's going wrong. Provide your command lines and Tesseract version you're using. Best r