Skip to content

Latest commit

 

History

3 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

Project Overview 📄🚀

This project is designed to automate the recognition of Khmer characters (consonants, vowels, and numerals). It addresses the complexity of the Khmer script, which includes unique features like subscript consonants (Cheung Akshar) and complex vowel placements.

✨ Key Components

1. Model Architecture:

  • It utilizes a Convolutional Neural Network (CNN), which is the standard for image classification tasks.
  • The model is trained to identify various classes of Khmer characters from image data.

2. Dataset:

  • The project uses a dataset of handwritten or digital Khmer characters.
  • Preprocessing involves resizing images (typically to 32x32 or 64x64 pixels) and normalizing pixel values to improve training efficiency.

3. Core Files:

🛠️ Features

  • High Accuracy: The model aims for a high recognition rate across different character types.
  • End-to-End Pipeline: Includes steps from data loading and augmentation (to handle different handwriting styles) to final prediction.
  • Keras/TensorFlow: The implementation is built using the TensorFlow framework and Keras API for deep learning.

📐 How it works

  • Input: An image of a single Khmer character.
  • Processing: The CNN extracts spatial features (curves, lines, and intersections) from the character.
  • Output: A classification label corresponding to the specific Khmer Unicode character (e.g., ក, ខ, គ).

Current Status: The repository serves as a research or educational tool for Khmer OCR (Optical Character Recognition).

About

A Deep Learning project using Convolutional Neural Networks (CNN) to classify and recognize Khmer script characters, including consonants and vowels.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages