Skip to content

Used MNIST and EMNIST dataset to deploy CNN to recognize handwritten characters using Tensorflow, Keras

Notifications You must be signed in to change notification settings

whysush/handwritten2text

Repository files navigation

Handwritten-to-Text OCR

This project uses TensorFlow and Keras to train and deploy convolutional neural network (CNN) models for identifying handwritten characters. The models are trained on the MNIST and EMNIST datasets to recognize both digits and alphanumeric characters.

Features

  • MNIST Model: Recognizes handwritten digits (0-9) from the MNIST dataset.
  • EMNIST Model: Identifies handwritten alphanumeric characters (letters and digits) from the EMNIST dataset.
  • Supports character recognition from uploaded images.

Files

  • mnist.ipynb: Digit identification using pre-set images from the MNIST dataset.
  • emnist.ipynb: Alphanumeric character identification with user-uploaded images (EMNIST dataset).

About

Used MNIST and EMNIST dataset to deploy CNN to recognize handwritten characters using Tensorflow, Keras

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published