tesseract

tesseract copied to clipboard

As the development of OCR is moving towards LSTM, new difficulties arise, such as having created lstm/traineddata for different fonts. **Example situations:** - I have created multiple lstm/ traineddata models,...

ghost

Font detection broken for segmention_mode > PSM_SINGLE_WORD

3

Font detection works fine in PSM_SINGLE_WORD mode. In PSM_SINGLE_LINE mode it is not working well. For recognizing more text (column, full page) font detection does not work at all. For...

taxophobia

How to prevent extra line breaks in text output?

8

Tesseract Version: 5 OS: LInux ubuntu 20 command: tesseract imagename --dpi 300 -l spa --psm 6 --oem 1 tried with different psm values, value 6 offers the best results **original...

info2000

How to join lines in a paragraph in text mode?

9

### Environment tesseract v5.0.0-alpha.20191030 Windows 64bit ### Current Behavior: When in text mode, each line in a paragraph is separated by line-break: ``` Line 1 of foo or bar, another...

spajak

Tesseract inserts extra blank lines

19

### Environment * **Tesseract Version**: 4.0.0 ~~4.0.0-beta.1 from https://packages.debian.org/stretch-backports/tesseract-ocr~~ * **Commit Number**: 51316994ccae0b48692d547030f26c0969308214 ~~c3ed6f036064e54e34f75275f66c70dd924527bf~~ * **Platform**: `Linux my-machine 4.9.0-8-amd64 #1 SMP Debian 4.9.130-2 (2018-10-27) x86_64 GNU/Linux` ### Current Behavior: Tesseract...

nezda

Sometimes failing to detect multiple columns

2

----------------------- ### Environment v5.0.0 alpha Windows 10-64 bit ### Current Behavior: I have two very similar documents seen here and here: In the former, Tesseract with psm=1 parses out two...

kuhanw

LSTM engine needs more training

5

According to Tesseract 4.0.0 [Release Notes](https://github.com/tesseract-ocr/tesseract/wiki/ReleaseNotes#tesseract-release-notes-oct-29-2018---v400) : > Added a new OCR engine that uses neural network system based on LSTMs, with major accuracy gains. My testing with this new...

YamashitaRen

Dropping words when trying with Telugu language

13

I tried to test with 4.0 with Telugu language and observed many words are dropping in between I given a 300 DPI PNG file.is this a known issue ? If...

harinath141

Issue with numbers recognition

14

Before you submit an issue, please review [the guidelines for this repository](https://github.com/tesseract-ocr/tesseract/blob/master/CONTRIBUTING.md). Please report an issue only for a BUG, not for asking questions. Note that it will be much...

nijanthan0

Text fragments of neighboring lines repeated

4

### Environment * **Tesseract Version**: Tesseract v5.0.0-alpha-20210401 * **Commit Number**: 38f0fdc * **Platform**: Linux/Debian ### Current Behavior: Repeats parts of preceding or following line. Looks like some memory constructs are...

wollmers