Tesseract OCR WebApp

Docker Hub - Jonesckevin/ocr-webapp

A Docker-based web application for OCR text extraction with AI-powered translation support. This app is meant to work offline using local LLMs (Ollama, LM Studio) with optional online cloud LLM support (OpenAI, Anthropic, Google Gemini, etc.). Great for your HALp needs!

Features

• OCR Processing: Extract text from images (PNG, JPG, TIFF, etc.) and PDF files
• Multi-language OCR: Support for 100+ languages with top 10 pre-installed
• AI Translation: Translate extracted text using local (Ollama, LM Studio) or cloud LLMs
• Multiple Upload Methods: Drag & drop, file select, or paste from clipboard
• File Management: Manage uploaded files and OCR results through the web interface
• Configurable: Customize titles, file limits, LLM settings via config file
• Secure API Keys: Server-side environment variables or client-side browser storage

Quick Start

Docker Hub

docker run -d -p 8202:5000 \
  -v ./data/uploaded:/data/uploaded \
  -v ./data/output:/data/output \
  -v ./config.json:/app/config.json:ro \
  jonesckevin/ocr-webapp

Using Docker Compose (Recommended)

1. Clone the repository:

   git clone https://github.com/Jonesckevin/OCR-WebApp.git
   cd OCR-WebApp

2. Start the application:

   docker compose up -d

3. Portainer:

   services:
     ocr-webapp:
       image: jonesckevin/ocr-webapp:latest
       container_name: ocr-webapp
       restart: unless-stopped
       ports:
         - "8202:5000"
       volumes:
         - ./data/uploaded:/data/uploaded
         - ./data/output:/data/output
         - ./data/logs:/data/logs
       environment:
         - FLASK_ENV=production
         - SECRET_KEY=change-this-to-a-secure-random-string
         - ALLOW_LIVE_LANG_INSTALL=true
         - ALLOW_CLIENT_API_KEYS=true
         - LOG_MAX_BYTES=10485760  # 10MB default
         - LOG_BACKUP_COUNT=5
         - LOG_LEVEL=INFO
         - OLLAMA_HOST=http://host.docker.internal:11434
         - LMSTUDIO_HOST=http://host.docker.internal:1234

4. Open your browser to http://localhost:8202

Building Manually

docker compose up --build
# or
docker build -t ocr-webapp .
docker run -d -p 8202:5000 \
  ocr-webapp

Configuration

Pre-installed OCR Languages

Priority languages (highlighted in UI):

• English (eng)
• French (fra)
• Ukrainian (ukr)
• Russian (rus)

Additional languages:

• German (deu)
• Spanish (spa)
• Italian (ita)
• Portuguese (por)
• Chinese Simplified (chi_sim)
• Japanese (jpn)

---

---

API Endpoints

Endpoint	Method	Description
/	GET	Main application page
/health	GET	Health check endpoint
/api/config	GET	Get application configuration
/api/progress/<task_id>	GET	Get progress for a specific task
/api/logs	GET	Get application logs
/api/logs/download	GET	Download complete log file
/api/logs/clear	POST	Clear log file
/api/upload	POST	Upload file
/api/upload/paste	POST	Upload pasted image (base64)
/api/files	GET	List uploaded files
/api/files/<filename>	GET	Download uploaded file
/api/files/<filename>	DELETE	Delete uploaded file
/api/languages	GET	List installed OCR languages
/api/languages/install	POST	Install new language pack
/api/ocr	POST	Perform OCR on a file
/api/results	GET	List OCR result files
/api/results/<result_id>	GET	Get specific OCR result
/api/results/<result_id>	DELETE	Delete OCR result
/api/results/<result_id>/edit	POST	Save edited OCR text
/api/results/<result_id>/download/<format>	GET	Download result (json/txt/translation)
/api/llm/test	POST	Test LLM provider connection
/api/llm/models	POST	Get available models for provider
/api/llm/translate	POST	Translate text using LLM
/static/<path:filename>	GET	Serve static files