--- title: Luna OCR emoji: 🌙 colorFrom: blue colorTo: purple sdk: static pinned: false license: mit thumbnail: https://b.l3n.co/i/NgyTV5.jpg tags: - ocr - text-extraction - ai - gemini - thai-language ---

Luna OCR

#

"Extract text from any image with AI precision... Don't let language barriers limit your potential, The world is full of text to discover."

--- ###

[FULLSCREEN](https://huggingface.co/veela4/luna_ocr/blob/main/README.md)


###

[🚀 Live Demo](https://huggingface.co/veela4/luna_ocr) | [📥 Download](https://huggingface.co/veela4/luna_ocr/tree/main) | [📋 Documentation](https://huggingface.co/veela4/luna_ocr/blob/main/README.md)


📋 Contents
#

- [🌙 What is Luna OCR?](#▼-preamble)

- [⚙️ Usage](#▼-usage)

- [🚀 Features](#▼-features)

- [🎨 Capabilities](#▼-capabilities)

- [📄 Document Processing 📋](#▼-documents) - [🇹🇭 Thai Language Support 🌏](#▼-thai) - [🖼️ Image Analysis 📸](#▼-images)

- [🏔️ Example Images ](#▼-examples) - [📄 Processing Demo 📋](#▼-processing) - [📄 Output Formats �](#▼-outputs)

- [📄 Instalelation](#▼-installation)

- [📝 Credits](#▼-credits)

- [License](#license)

- [Disclaimer](#disclaimer)


🌙 What is Luna OCR?

**Luna OCR** is an AI-powered text extraction application that transforms images into editable text with unprecedented accuracy. Built with Google's Gemini 2.5 Flash model, Luna specializes in multilingual text recognition with particular excellence in Thai language processing. The name Luna, meaning 'moon' in Latin, represents illumination in darkness - bringing clarity to text hidden within images.


⚙️ Usage

##

Simple steps to extract text from any image: | Step | Action | | ----------------- | ------------------------------------------------------------------ | | 1 | Upload your image (PNG, JPG, PDF supported) | | 2 | Enter your Gemini API key (stored securely locally) | | 3 | Click "Process Image" | | 4 | Get extracted text with formatting preserved | | 5 | Copy, edit, or download results | | Optional | Enable Thai language optimization | **🔑 API Key Setup:** Get your free Gemini API key from [Google AI Studio](https://makersuite.google.com/app/apikey)

Luna Processing Interface

🚀 Features

##

Core capabilities that make Luna OCR special:

**

Key Features:**

✅ **AI-Powered Accuracy** - Gemini 2.5 Flash model for superior text recognition ✅ **Multi-Language Support** - Optimized for Thai, English, and 100+ languages ✅ **Privacy First** - All processing happens locally, your images never leave your device ✅ **Format Preservation** - Maintains original text structure and formatting ✅ **Multiple File Types** - Supports PNG, JPG, JPEG, PDF, and more ✅ **Real-time Processing** - Fast extraction with live progress indicators ✅ **Responsive Design** - Works perfectly on desktop and mobile devices


🎨 Capabilities

## Luna OCR excels in various document processing scenarios..

You can achieve professional results with all these examples.


Document Processing 📄

**Perfect for:** - Business documents - Academic papers - Legal contracts - Technical manuals - Multi-page PDFs **Features:** - Page-by-page processing - Table structure preservation - Header/footer recognition - Formatting maintenance # **
Luna OCR
**


Thai Language Support 🇹🇭

**Specialized for Thai:** - Proper word spacing - Context-aware processing - Cultural text understanding - Mixed Thai-English support **Advanced Features:** - Tone mark recognition - Complex character handling - Traditional script support - Modern typography processing # **
Luna OCR
**


Image Analysis 📸

**Image Types:** - Screenshots - Photographs - Scanned documents - Digital artwork with text - Social media posts **Processing Power:** - Auto-enhancement - Noise reduction - Rotation correction - Multi-format support # **
Luna OCR
**



🏔️ Example Images
Processing Demo 📄 PDF Processing Demo

**Input:** Multi-page PDF documents **Output:** Structured text with preserved formatting **Features:** Table recognition, header/footer handling # **LUNA OCR**


Output Formats � Text Output **Available Formats:** - Plain Text (clean, readable) - Markdown (structured with tables) - JSON (structured data) - HTML (styled output) **Export Options:** - Copy to clipboard - Download as file - Direct editing # **LUNA OCR**

💡 Installation

## Quick Start Guide ```bash # Clone the repository git clone https://huggingface.co/veela4/luna_ocr cd luna_ocr # Install dependencies npm install cd server && npm install && cd .. # Start the application npm start ``` ## Requirements - Node.js 18+ - NPM or Yarn - Modern web browser - Gemini API key (free from Google AI Studio) ## Access Points - **Frontend:** http://localhost:3000 - **Backend:** http://localhost:3002 - **Health Check:** http://localhost:3002/api/health


📝 Credits

Luna OCR is built with cutting-edge technologies and open-source libraries.

**AI Model:** Google Gemini 2.5 - Advanced multimodal AI

**Frontend:** React 19 - Modern UI framework

**Backend:** Node.js + Express - Server infrastructure

**3D Graphics:** Three.js - WebGL animations

**UI Components:** Radix UI - Accessible components

**Animations:** Framer Motion - Smooth transitions

**Special thanks to the open-source community for making this project possible.**


# License This application is open source and available under the MIT License. The MIT License specifies: ``` 1. You can use, copy, modify but aren't subject to any illegal activites or selling this product 2. The software is provided "as is", without warranty of any kind 3. Authors are not liable for any claim, damages or other liability 4. You must include the original copyright notice and license in all copies ``` [Please read the full license here](https://opensource.org/licenses/MIT) You're free to use Luna OCR for License purpose, for commercial applications please contact.
| Allowed | Permission | |:-------:|-----------------------------------------------------| | 🌙 | Use for personal projects | | 🌙 | Use for commercial applications | | 🌙 | Modify and distribute | | 🌙 | Create derivative works | | 🌙 | Private use | | 🌙 | Patent use | # Disclaimer ``` Luna OCR processes images locally and does not store any user data. API keys are encrypted and stored locally on your device. The application requires a Gemini API key for text processing functionality. Users are responsible for complying with applicable laws and regulations when processing documents. ```