Handwriting Recognition with Flask and Microsoft T
W A lightweight Flask web app that uses Microsoft’s TrOCR model from Hugging Face Transformers to recognize and extract text from uploaded handwritten images. Simply upload an image, and the app processes it to display the recognized handwriting instantly.
Project Summery
A simple yet powerful web application that performs handwriting recognition using Microsoft’s TrOCR model and displays the recognized text from uploaded handwritten image files.
Built with Flask (Python web framework) and Transformers (by Hugging Face).
🚀 Features
- Upload
.jpg,.jpeg,.pnghandwritten image files - Recognizes handwritten text using TrOCR (Transformer OCR)
- Displays extracted text in a clean web interface
- Works offline after initial model download
🧰 Tech Stack
| Layer | Tech |
|---|---|
| Backend | Python, Flask |
| AI Model | microsoft/trocr-base-handwritten (Hugging Face) |
| Frontend | HTML, CSS (basic) |
| Image Processing | Pillow (PIL) |
📦 Installation
1. Clone the Repository
git clone https://github.com/sujit-codezen/Handwriting-Recognition-Web-App.git cd handwriting-recognizer
2. Create a Virtual Environment
python -m venv venv source venv/bin/activate # on Windows: venv\Scripts\activate
3. Install Requirements
pip install -r requirements.txt
4. Run the App
python app.py
5. Output
