node-paddle-ocr

Offline, Multi-language and Faster OCR in Terminal.

---

English | 简体中文

An OCR terminal tool based on the open-source offline components of PaddleOCR v2.6 C++.

1. Clone this repo
2. Run cd node-paddle-ocr
3. Run npm run build
4. Run npm link

ocr <image path> [options]

Options:
  -t, --target   Detection language specified as ISO 639-1 code         [string]
  -c, --code     List supported language codes                         [boolean]
  -v, --version  Show version number                                   [boolean]
  -h, --help     Show help                                             [boolean]

Examples:
  $ ocr              Identify pictures in the clipboard
  $ ocr ./image.png  Identify the image.png file in the current directory

• The system supports Win10/11 x64.
• It is not recommended to use Win7 as the recognition engine is highly likely to be unable to run.
• The CPU must have the AVX instruction set. Common home-use CPUs generally meet this condition.

• PaddleOCR-json - An OCR offline image text recognition command-line Windows program that outputs results in the form of JSON strings for convenient invocation by other programs. It provides APIs in various languages. Compiled by PaddleOCR C++.
• PaddleOCR-json-node-api - A node.js API based on PaddleOCR-json.

Contributions via Pull Requests or Issues are welcome.

This project is licensed under the MIT License. See the LICENSE file for details.