[tesseract-ocr] Re: tesseract performs wrong auto-correction sometimes : how to disable it?

I am experiencing the same issue. Did you ever find a resolution for this? On Wednesday, 25 April 2018 10:59:34 UTC-4, Youcef wrote: > > Hi, > > > Tesseract seems to post process its prediction. > > Here after, what I get after OCRizing images (same font, same size images > generated with text2i

[tesseract-ocr] Errors when numeric and alphabetic data is mixed

I am using the API to read data from an image. I have created training files for the fonts I process and I pre-process the image to deskew and clean it. When I read entirely numeric data it reads perfectly e.g. 123456. When I read entirely alphabetic data it reads perfectly e.g. ABCDEFGH. The pro

[tesseract-ocr] Tesseract misreading numbers

I am using Tesseract 3.0.5 with the .Net wrapper installed from the Nuget package https://www.nuget.org/packages/Tesseract/ . I have a program which reads text from an image. I am using PSM = 6 and pre-process the image to remove noise. I select a portion of the image for processing and have

[tesseract-ocr] Accuracy decreases when a Region of Interest is used

I am using Tesseract 3.0.4 with the Tesseract-OCR .Net wrapper. I am reading a page which contains account numbers and payment amounts along with other data. If I read the entire page using ... var Page2 = Engine.Process(PixPage, PageSegMode.Auto) , the account numbers and payment amounts are