https://github.com/nafisalawalidris/building-a-cnn-to-classify-handwritten-characters

Building a CNN to Classify Handwritten Characters" is a proof-of-concept project using Python, TensorFlow, and Keras. It demonstrates how a Convolutional Neural Network (CNN) can identify scanned characters in the MNIST database. The goal is to convert historical documents into searchable databases, improving ancestry search services.
https://github.com/nafisalawalidris/building-a-cnn-to-classify-handwritten-characters

data-science database handwritten-characters image-recognition jupyter-notebook keras machine-learning proof-of-concept python tensorflow

Last synced: 3 days ago
JSON representation

Building a CNN to Classify Handwritten Characters" is a proof-of-concept project using Python, TensorFlow, and Keras. It demonstrates how a Convolutional Neural Network (CNN) can identify scanned characters in the MNIST database. The goal is to convert historical documents into searchable databases, improving ancestry search services.

• Host: GitHub
• URL: https://github.com/nafisalawalidris/building-a-cnn-to-classify-handwritten-characters
• Owner: nafisalawalidris
• Created: 2023-07-23T17:25:37.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-03-23T21:28:04.000Z (10 months ago)
• Last Synced: 2024-11-22T21:16:28.933Z (2 months ago)
• Topics: data-science, database, handwritten-characters, image-recognition, jupyter-notebook, keras, machine-learning, proof-of-concept, python, tensorflow
• Language: Jupyter Notebook
• Homepage:
• Size: 94.7 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Building a Convolutional Neural Networks (CNN) with TensorFlow to Classify Handwritten Characters

A proof-of-concept project for converting scanned historical documents into searchable databases using CNNs.

## Scenario

A company providing ancestry search services needs to process handwritten documents. This project demonstrates how a CNN can identify scanned characters in the MNIST database with 60,000 training samples and 10,000 test samples.

![Image Description](https://raw.githubusercontent.com/nafisalawalidris/Building-a-CNN-to-Classify-Handwritten-Characters/main/the%20model's%20predictions.png)

### Tools and Technologies:

- Python

- TensorFlow

- Keras

- Jupyter Notebook

- Machine Learning

- Image Recognition

- Data Science

### Here's the link to the NIST Special Database 19, which contains a large collection of handwritten characters:

[NIST Special Database 19](https://www.nist.gov/srd/nist-special-database-19)

You can visit the provided link to access the dataset and find more information about it on the NIST website.

### Project Completion:

Feel free to check out the complete project on my GitHub repository!

Link: [GitHub Repository](https://github.com/elfeenah/Building-a-CNN-to-Classify-Handwritten-Characters/edit/main/README.md)

### Contributions:

Contributions to this project are welcome! If you have any suggestions, improvements, or bug fixes, please submit a pull request. Let's collaborate and make this project even better together!