title: Hand Written Text Digit Recognition
emoji: π
colorFrom: blue
colorTo: red
sdk: streamlit
sdk_version: 1.28.1
app_file: app.py
pinned: false
license: mit
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
Task : Handwritten Text Digit Recognition using streamlit
Problem Statement:
Handwritten digit recognition is a fundamental task in machine learning and computer vision that involves identifying and classifying individual digits (0-9) that have been handwritten. This task plays a crucial role in various applications across different domains. This project aims to develop a handwritten digit recognition system using deep learning techniques. It involves building and training models that can accurately classify handwritten digits.
Importance and Use Cases:
Finance: Automatic check reading, where handwritten numbers on checks are recognized to process financial transactions efficiently.
Postal Services: ZIP code recognition, aiding postal services in sorting and delivering mail accurately.
Data Entry and Forms: Digit extraction in forms, simplifying data entry processes and reducing errors in handwritten information.
Educational Tools: Handwritten digit recognition is also used in educational tools to provide interactive learning experiences, such as math and number games.
OCR Systems: It is a crucial component of Optical Character Recognition (OCR) systems used for digitizing printed and handwritten text.
Challenges:
Handwritten digit recognition presents several challenges, including variations in writing styles, different writing tools, noise in scanned images, and variations in digit orientation and size.
Overview
This project aims to implement a Handwritten Digit Recognition system. We have trained a machine learning model and loaded the model weights (model_weights.pth). By running the app.py script, you can interact with the application.
Usage Instructions
Follow these steps to use the application:
Installation: Ensure you have installed all the required libraries and packages.
Model Setup: We've trained a model and saved its weights in model_weights.pth. Make sure you have this file in your project directory.
Run the App: Execute the app.py script to launch the interactive application.
Select Colors: Use the application to choose the background color and stroke color. You can also load an input image or draw a digit.
Digit Prediction: The model will predict the drawn digit or the digit in the loaded image.
Enjoy using the Handwritten Digit Recognition app!