SandhyaPitchika/Hindi-Indian-Sign-language-dataset-ISL

Here is a professionally formatted README.md file designed for your Hugging Face dataset. This file is optimized to help you rank on search engines while clearly linking back to your Kaggle repository.

---

Indian Sign Language: Hindi (ISL-Hindi)

India’s First Comprehensive Image Dataset for Hindi Character Recognition

---

Overview

This dataset is the first of its kind, specifically curated to address the lack of resources for regional Indian languages in the field of Computer Vision and Accessibility. It contains high-quality, hand-captured images of Indian Sign Language (ISL) gestures representing the Hindi alphabet (Devanagari script).

• Primary Source (Kaggle): Link to Kaggle Dataset
• Creator: [Your Name/Username]
• Total Images: 48,000
• Classes: 40 (Vowels, Consonants, and Compound characters)

---

Dataset Structure

The data is organized into a clean, categorical folder structure. Every character folder contains exactly 1,200 images, ensuring a perfectly balanced dataset for model training.

HindiSignImages48x48/
├── अ/ (1200 images)
├── आ/ (1200 images)
├── ... [36 intermediate folders]
├── क्ष/ (1200 images)
└── ज्ञ/ (1200 images)

Technical Specifications

• Resolution: 48x48 pixels (optimized for CNNs).
• Format: RGB/Grayscale images in .jpg format.
• Balance: Perfectly balanced (1,200 samples per class).

---

Provenance & Collection

As the creator, I captured and processed this dataset entirely from scratch to ensure the highest data integrity:

1. Manual Capture: Recorded using high-definition equipment under varied lighting and background conditions.
2. Hand-Cropped: Every image was manually cropped to isolate the hand gesture and remove background noise.
3. India's First: This is an original primary-source dataset, not a web-scraped collection.

---

How to Use

You can use this dataset directly via the Hugging Face datasets library or by downloading it from the Kaggle link provided above.

# Coming soon: load_dataset integration
# For now, please download and use the ImageFolder structure

---

---

External Resources

• Kaggle Dataset: Indian Sign Language Hindi
• Project Goal: To provide the first standardized resource for Hindi ISL character recognition.

---

Citation & License

If you use this dataset in your research or project, please cite the primary Kaggle source.

• License: [Insert License, e.g., CC BY-SA 4.0]