Switch to full style
Post all Gimp scripts and script writing questions here
Post a reply

Script OCR Image (Tesseract OCR engine )

Sun Jun 07, 2020 11:38 pm

BubbleOCR 2.2 (Balloon Text)
NEW VERSIONS IN
https://github. com/snakeotakon/BubbleOCR

This script allows you to perform text recognition of the text balloons of an image, it also allows you to translate the text using Google Translator.
It can be translated from any language, Japanese, Korean, Chinese to English, Spanish. Russian.
Use the Engine Tesseract (CaptureText), to recognize the text.
It can be used for multiple images at once,

In videos, link all files necessarys for work
All Files necessary waifu2x,engine tesseract,tessdata, 86_Redistributable for Visual Studio 2015 (480 MB)


30/06/2020
Add selection automatic (detection text for bubbles)
13/09/2022
BubbleOCR 2.2

Include other scripts
Script_SelectionToPath.py.- convert selection to path for work with OCR Ballon text
Script_SelectionToPath_Inverse.py.- convert "selection to path" or "path to selection", for corrections
Script_ShowLayer.py
Script_Waifu2x.py.- Scale with Waifu2x
Script_ChangeFont.py
Script_FilterPersonalized.py.- set any function for layers selected
Script_OCRForBalloonsText.py
Script_OrderLayers.py
Script_ParentGroupForLayers.py
Script_QuickViewRemoveHoles.py
Script_RemoveLayers.py
Script_RepeatKey.py.- repeat any filter with shortcut (Ctrl+F), for layers selected

Tesseract neural net (LSTM) based OCR engine

Re: Script OCR Image (Tesseract OCR engine )

Tue Jun 09, 2020 8:39 pm

This looks like a competent python script.
I guess the reason there is little interest here is because people rarely want to know what's in a foreign language speech balloon.

I've often wondered if North Koreans even have cartoons? Apparently no.

Re: Script OCR Image (Tesseract OCR engine )

Sat May 22, 2021 9:09 pm

great work, i have been searching for such solution for years and it really worked.
thanks so much for such effort, but i have some errors:
1. the google chrome cant open because i have x64 version of chrome and i cant install x84 version.
2. the google translate token gives error , i guess its because of token

Re: Script OCR Image (Tesseract OCR engine )

Sun May 23, 2021 3:57 am

i guess that googletrans library need upgrade like mentioned in stackoverflow question 52455774, as im not allowed to post urls

Re: Script OCR Image (Tesseract OCR engine )

Tue Sep 13, 2022 11:24 pm

New Version Bubble OCR 2.2

github. com/snakeotakon/BubbleOCR

-Substitution Lists
-Compatible Edge
-Fix Bugs

Re: Script OCR Image (Tesseract OCR engine )

Wed Sep 14, 2022 1:26 am

https://github.com/snakeotakon/BubbleOCR
Post a reply