📝 OCR API with FastAPI and Tesseract

This project is a FastAPI-based OCR (Optical Character Recognition) API that uses Tesseract-OCR to extract text from images. The API is containerized with Docker, deployed using Portainer, and the image is stored in AWS ECR.

---

🚀 Features

• ✅ FastAPI framework for high-performance API.
• ✅ Tesseract-OCR for optical character recognition.
• ✅ Pre-processing techniques (grayscale, contrast adjustment, binarization).
• ✅ Dockerized application for easy deployment.
• ✅ AWS ECR for container storage.
• ✅ Portainer compatibility for easy management.
• ✅ Optimized Tesseract settings (--oem 1 --psm 3 -l ita).

---

🏗️ Project Structure

📦 ocr-tesseract ┣ 📂 app ┃┣ 📜 main.py # FastAPI application ┃┗ 📜 ocr.py # OCR processing logic ┣ 📂 infrastructure ┃ ┣ 📜 Dockerfile # Docker setup ┃ ┣ 📜 docker-compose.yml # Container configuration ┃ ┗ 📜 requirements.txt # Python dependencies ┗ 📜 README.md

---

🔧 Installation & Setup

1️⃣ Clone the repository

2️⃣ Set up a virtual environment (optional)

source venv/bin/activate  # On Windows use: venv\Scripts\activate
pip install -r infrastructure/requirements.txt

3️⃣ Run the API Locally with Docker Compose

Instead of running manually, use Docker Compose:

cd infrastructure
docker-compose up --build -d

This will:

Build the Docker image Start the container in detached mode (-d) Expose the API on port 8500