Re: Reading image with few digits

Check this thread https://groups.google.com/forum/?fromgroups#!topic/tesseract-ocr/TY_RIHOOyNM Read about the "psm" switch and custom segmentation. Likely these can help you Warm regards, Dmitri Silaev www.CustomOCR.com On Tue, Feb 14, 2012 at 11:42 AM, ReneFR wrote: > Tesseract returns "empty

Re: Reading image with few digits

Tesseract is designed for whole pages and needs context. Please read the archives of this list for details. Duplication and post processing works, as does segmentation. Chech out opencv. --Sven On Tuesday, February 14, 2012, ReneFR wrote: > Tesseract returns "empty page" when i submit a image wi

Reading image with few digits

Tesseract returns "empty page" when i submit a image with only few digits (1,2 or 3...) for exemple : a png file with 111 return "empty page" but works fine. or 8.8 failed (empty page) but 888 works. I'm using the 3.01 version (same effect with 3.00...) Thanks for your help. -- You receive