riotu-lab/SARD-Extended

SARD: Synthetic Arabic Recognition Dataset

 

Overview

SARD (Synthetic Arabic Recognition Dataset) is a large-scale, synthetically generated dataset designed for training and evaluating Optical Character Recognition (OCR) models for Arabic text. This dataset addresses the critical need for comprehensive Arabic text recognition resources by providing controlled, diverse, and scalable training data that simulates real-world book layouts.

Key Features

• —Massive Scale: 2,621,075 document images containing 794.6 million words
• —Extensive Typographic Diversity: Ten distinct Arabic fonts covering a wide range of styles
• —Structured Formatting: Designed to mimic real-world book layouts with consistent typography
• —Clean Data: Synthetically generated with no scanning artifacts, blur, or distortions
• —Content Diversity: Text spans multiple domains including culture, literature, Shariah, social topics, and more

Dataset Structure

The dataset is divided into ten splits based on font name:

• —Amiri: Classical Naskh typeface inspired by early 20th century typography
• —Sakkal Majalla: Widely used font in contemporary Arabic publishing
• —Arial: Modern sans-serif font common in digital publications
• —Calibri: Microsoft's default font representing contemporary digital typography
• —Scheherazade New: Traditional-style font based on classical manuscript styles
• —Jozoor Font: Decorative Arabic font with more stylized character forms
• —Lateef: Extended Arabic script font designed for readability at small sizes
• —Noto Naskh Arabic UI: Part of Google's Noto family, designed for user interfaces
• —Thabit: Monospaced Arabic font for technical documentation
• —Al Jazeera Arabic Regular: Based on the typography used by Al Jazeera media
• —Traditional Arabic: Is a TrueType Arabic font with a regular style and installable embedding

📋 Sample Images <div align="center"> <table> <tr> <td><img src="https://cdn-uploads.huggingface.co/production/uploads/64e8eb21233101ed99b204c8/gwF9jkkkpzRSzrP9GCEl.png" width="300" alt="Sample 1 - Amiri Font"/></td> <td><img src="https://cdn-uploads.huggingface.co/production/uploads/64e8eb21233101ed99b204c8/dsqWoCh5x31eGq-u-PqPS.png" width="300" alt="Sample 2 - Arial Font"/></td> </tr> <tr> <td><img src="https://cdn-uploads.huggingface.co/production/uploads/64e8eb21233101ed99b204c8/2XK9Ey6k6HSDXKXCxmVRG.png" width="300" alt="Sample 3 - Calibri Font"/></td> <td><img src="https://cdn-uploads.huggingface.co/production/uploads/64e8eb21233101ed99b204c8/CxKITKvc3EnDIuqnNybV.png" width="300" alt="Sample 4 - Scheherazade Font"/></td> </tr> </table> </div>

Each split contains data specific to a single font with the following attributes:

• —image_name: Unique identifier for each image
• —chunk: The text content associated with the image
• —font_name: The font used in text rendering
• —image_base64: Base64-encoded image representation
• —sample_id: Unique ID
• —article_link: Link of the source article

Content Distribution

Font Specifications

Page Layout

Usage Example

from datasets import load_dataset
import base64
from io import BytesIO
from PIL import Image
import matplotlib.pyplot as plt

# Load dataset with streaming enabled
ds = load_dataset("riotu-lab/SARD-Extended", streaming=True)
print(ds)

# Iterate over a specific font dataset (e.g., Amiri)
for sample in ds["Amiri"]:
    image_name = sample["image_name"]
    chunk = sample["chunk"]  # Arabic text transcription
    font_name = sample["font_name"]
    
    # Decode Base64 image
    image_data = base64.b64decode(sample["image_base64"])
    image = Image.open(BytesIO(image_data))

    # Display the image
    plt.figure(figsize=(10, 10))
    plt.imshow(image)
    plt.axis('off')
    plt.title(f"Font: {font_name}")
    plt.show()

    # Print the details
    print(f"Image Name: {image_name}")
    print(f"Font Name: {font_name}")
    print(f"Text Chunk: {chunk}")
    
    # Break after one sample for testing
    break

Working with Multiple Fonts

To train or evaluate models across different font styles:

from datasets import load_dataset
import random

# Load the dataset
ds = load_dataset("riotu-lab/SARD")

# Select a balanced sample from multiple fonts
fonts_to_use = ["Amiri", "Arial", "Scheherazade_New", "Thabit", "Noto_Naskh_Arabic_UI"]
samples_per_font = 1000
combined_samples = []

for font in fonts_to_use:
    # Get random samples from this font
    font_samples = ds[font].shuffle(seed=42).select(range(samples_per_font))
    combined_samples.extend([(sample, font) for sample in font_samples])

# Shuffle the combined samples
random.shuffle(combined_samples)

# Now you can use these samples for training or evaluation
for sample, font in combined_samples[:5]:  # Just show first 5 as example
    print(f"Font: {font}, Image: {sample['image_name']}")

Applications

SAND is designed to support various Arabic text recognition tasks:

• —Training and evaluating OCR models for Arabic text
• —Developing vision-language models for document understanding
• —Fine-tuning existing OCR models for better Arabic script recognition
• —Benchmarking OCR performance across different fonts and layouts
• —Research in Arabic natural language processing and computer vision
• —Developing font-adaptive OCR systems that generalize across typographic styles

Acknowledgments

The authors thank Prince Sultan University for their support in developing this dataset.

Citation

If you use SARD in your work, please cite the following paper:

APA:

Nacar, O., Al-Habashi, Y., Sibaee, S., Ammar, A., & Boulila, W. (2025). SARD: A Large-Scale Synthetic Arabic OCR Dataset for Book-Style Text Recognition. arXiv preprint arXiv:2505.24600.
https://arxiv.org/abs/2505.24600

BibTeX:

@misc{nacar2025sard,
      title={SARD: A Large-Scale Synthetic Arabic OCR Dataset for Book-Style Text Recognition}, 
      author={Omer Nacar and Yasser Al-Habashi and Serry Sibaee and Adel Ammar and Wadii Boulila},
      year={2025},
      eprint={2505.24600},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2505.24600}, 
}

Usage Note:

This dataset includes text sourced from Alukah. Use is limited to non-commercial purposes in accordance with the source terms. Redistribution, republication, or commercial use is not permitted without permission and proper attribution to the original source and author.