https://github.com/gnonio/korporize

OCR - Object Character Recognition for any image you browse upon
https://github.com/gnonio/korporize

javascript ocr-recognition webextensions

Last synced: 14 days ago
JSON representation

OCR - Object Character Recognition for any image you browse upon

• Host: GitHub
• URL: https://github.com/gnonio/korporize
• Owner: gnonio
• License: apache-2.0
• Created: 2020-05-06T21:46:42.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-05-20T22:28:25.000Z (over 4 years ago)
• Last Synced: 2024-08-01T16:44:46.372Z (3 months ago)
• Topics: javascript, ocr-recognition, webextensions
• Language: JavaScript
• Homepage:
• Size: 35.5 MB
• Stars: 11
• Watchers: 1
• Forks: 2
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
[![korporize](./img/korporize.svg)](http://tesseract.projectnaptha.com)

## korpora - OCR - Optical Character Recognition

Offline text recognition from any image. This web extension will enable context menu access to extract text from any image while browsing. Builds upon [Tesseract.js](https://github.com/naptha/tesseract.js)

****

### Install

- [Addons Mozilla page](https://addons.mozilla.org/en-GB/firefox/addon/korporize/)

#### Alternate install for advanced users

- [download this repository](https://github.com/gnonio/korporize/archive/master.zip)

- follow instructions for [Temporary installation in Firefox](./user-install.md)

****

### Usage

- Right click over an image in a web page

- Select "Extract Text from Image"

- A popup will open with korporize interface

- Wait for tesseract to work in the background

- Obtain results in korporize panel

- (Optional) copy results to clipboard

****

To obtain good results:

- make sure the automatic language detected is suitable for the characters in the image loaded

- force another language via Options page

- increase quality in Options page

(try Normal or Best - both will take longer)

- make sure you have a suitable page segmentation for the image

(will make this choice handier in future releases)

- choose a high resolution version of the image

****

### Features

- Extracts text from any image while browsing

- Works offline (requires network only the first time a language is used to cache the dictionaries)

- Automatic language detection (based on the visited web page)

- Prevents downloading twice already loaded images

****

### Notes

- Careful with the size of language dictionaries

- Expect around 8Mb for Normal and 12Mb for Best Quality per language

- Aside from above dictionaries no other data is ever stored by korporize

****

### Todo

- Many other options for accessing Tesseract functionality (image from link, PDF load and save, etc...)

- Preloading of language dictionaries (via Options page)

- Provide some cache management options

- Provide access as an API for other webextensions