Skip to content

afreenAI/Digit-Recognition-System

Repository files navigation

Digit-Recognition-System

CNN-based multi-digit recognition using deep learning

Digit Recognition System using Deep Learning

This project implements a CNN-based digit recognition system capable of recognizing single and multi-digit numbers from real-world images.

Features

  • CNN trained on MNIST dataset
  • Single-digit and multi-digit recognition
  • Robust image preprocessing
  • Digit segmentation and ordering
  • Confidence score for predictions

Technologies Used

  • Python
  • TensorFlow / Keras
  • OpenCV
  • NumPy

Applications

  • OTP recognition
  • Roll number extraction
  • PIN / postal code reading
  • Cheque number recognition

How to Run

  1. Install dependencies: pip install -r requirements.txt
  2. Run the script: python multi_digit_recognition.py

Note

The model is trained using the MNIST dataset and tested on real-world images.

About

CNN-based multi-digit recognition using deep learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages