maddy2104/Sindhi-Handwritten-Alphabets

This dataset is a rich collection of handwritten Sindhi alphabet images, carefully curated to capture a diverse range of writing styles. The dataset includes samples from multiple generations, including Gen X, Millennials, Gen Z, and Gen Alpha, ensuring a broad representation of handwriting variations. Additionally, it encompasses contributions from individuals of different genders and varying levels of handwriting proficiency, making it highly valuable for research in handwriting recognition and computer vision.

This dataset is ideal for training machine learning models on tasks such as:

• Optical Character Recognition (OCR) for Sindhi script
• Handwriting style analysis across generations
• Character classification and dataset augmentation experiments
• Computer vision research involving regional scripts

Dataset Details:

• Total Sindhi Letters: 52
• Images per Letter: 58
• Total Images: 3016
• Format: JPG
• Diversity Factors:
  • Generations Covered: Gen X, Millennials, Gen Z, Gen Alpha
  • Gender Representation: Male and Female contributors
  • Writing Styles: Includes cursive, bold, thin, and uneven strokes

Dataset Structure:

The dataset is structured into 52 folders, each representing a unique Sindhi letter. Each folder contains 31 handwritten samples of that letter, captured from various contributors.

Usage & Applications:

This dataset can be used by researchers, educators, and developers working on:

• Handwriting Recognition Models
• AI-powered OCR for Sindhi Language
• Multi-generational Handwriting Studies
• Sindhi Language Digitization & Preservation

License & Attribution

This dataset is publicly available under the CC BY 4.0 License, meaning you can use it freely with proper attribution.

This dataset was created through the combined efforts of multiple contributors who provided handwritten samples.

• Mudasir Murtaza
• Farhad Ali

This dataset is now open-source and we encourage researchers, developers, and students to use this dataset for AI projects and Sindhi handwriting recognition models!