Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -1,15 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
|
2 |
-
## Dataset Details
|
3 |
|
4 |
-
- **Total Images**: 36954
|
5 |
-
- **Font Size**: [14, 16, 18, 20, 24, 32, 40, 60, 80, 100]
|
6 |
-
- **Page Layout**: {'A4': {'size': '210mm 297mm'}, 'Letter': {'size': '216mm 279mm'}, 'Small': {'size': '105mm 148mm'}, 'Square': {'size': '1080px 1080px'}, 'OneLine': {'size': '210mm 10mm'}}
|
7 |
-
- **Font Used**: ['fonts/arabic/IBMPlexSansArabic-Regular.ttf', 'fonts/arabic/KFGQPCUthmanTahaNaskh.ttf', 'fonts/arabic/ScheherazadeNew.ttf', 'fonts/arabic/madina.otf', 'fonts/arabic/Amiri.ttf', 'fonts/arabic/DiwaniLetter.ttf', 'fonts/arabic/Tajawal-Regular.ttf', 'fonts/arabic/Cairo.ttf', 'fonts/arabic/Lateef.ttf', 'fonts/arabic/Almarai-Regular.ttf', 'fonts/arabic/AlQalamQuran.ttf', 'fonts/arabic/NotoNaskhArabic.ttf', 'fonts/arabic/NotoNaskhArabic-Regular.ttf']
|
8 |
-
- **Datasets Used**:
|
9 |
-
- [YoussefAnwar/Arabic-news](https://huggingface.co/datasets/YoussefAnwar/Arabic-news) (max 40000 samples from 'train' split, using 'Body' column)
|
10 |
-
- [oddadmix/Basma2423-Text-with-Diacritics-Correction](https://huggingface.co/datasets/oddadmix/Basma2423-Text-with-Diacritics-Correction) (max 40000 samples from 'train' split, using 'vocalized' column)
|
11 |
|
12 |
-
## License
|
13 |
|
14 |
-
The dataset is released under the MIT License.
|
15 |
-
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
task_categories:
|
4 |
+
- text-to-image
|
5 |
+
language:
|
6 |
+
- ar
|
7 |
+
tags:
|
8 |
+
- arabic
|
9 |
+
- Qari
|
10 |
+
- OCR
|
11 |
+
- ArabicOCR
|
12 |
+
- BookStyle
|
13 |
+
- Markdown
|
14 |
+
pretty_name: Qari-OCR
|
15 |
+
size_categories:
|
16 |
+
- 10K<n<100K
|
17 |
+
---
|
18 |
+
|
19 |
+
# QARI Markdown Mixed Dataset
|
20 |
+
|
21 |
+
<div align="center">
|
22 |
+
|
23 |
+

|
24 |
+

|
25 |
+

|
26 |
+

|
27 |
+
|
28 |
+
</div>
|
29 |
+
|
30 |
+
<div align="center">
|
31 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/628f7a71dd993507cfcbe587/Txz_HjVy6NsdmcghXqVH_.png" alt="QARI Logo" width="400">
|
32 |
+
</div>
|
33 |
+
|
34 |
+
## 📋 Dataset Summary
|
35 |
+
|
36 |
+
The QARI Markdown Mixed Dataset is a specialized synthetic dataset designed for training Arabic OCR models with a focus on complex document layouts and HTML structure understanding.
|
37 |
+
This dataset is part of the QARI-OCR project, which achieves state-of-the-art performance in Arabic text recognition.
|
38 |
+
|
39 |
+
This dataset contains **50,000 synthetically generated Arabic document images** with corresponding ground truth text in HTML/Markdown format, featuring:
|
40 |
+
- 🔤 **Full diacritical marks (tashkeel)** support
|
41 |
+
- 📝 **Mixed font sizes** within documents (headers, body text, annotations)
|
42 |
+
- 🎨 **12 distinct Arabic fonts** ranging from common Naskh to ornate calligraphic styles
|
43 |
+
- 📄 **Realistic document layouts** with structural HTML tags
|
44 |
+
- 🖼️ **Three degradation levels**: Clean, Moderate, and Heavy
|
45 |
+
|
46 |
+
## 🎯 Intended Use
|
47 |
+
|
48 |
+
This dataset is specifically designed for:
|
49 |
+
- Training OCR models that need to understand document structure
|
50 |
+
- Fine-tuning vision-language models for Arabic text recognition
|
51 |
+
- Developing systems that preserve formatting and layout information
|
52 |
+
- Research in Arabic document analysis and understanding
|
53 |
+
|
54 |
+
## 📊 Dataset Statistics
|
55 |
+
|
56 |
+
| Metric | Value |
|
57 |
+
|--------|-------|
|
58 |
+
| **Total Images** | 50,000 |
|
59 |
+
| **Text Sources** | Modern news articles + Classical Islamic corpus |
|
60 |
+
| **Font Variety** | 12 Arabic fonts |
|
61 |
+
| **Font Size Range** | 14px - 100px |
|
62 |
+
| **Degradation Types** | 3 (Clean, Moderate, Heavy) |
|
63 |
+
| **Diacritics Support** | ✅ Full tashkeel |
|
64 |
+
| **HTML Structure** | ✅ Preserved |
|
65 |
+
| **Layout Complexity** | ✅ High (mixed sizes, headers) |
|
66 |
+
|
67 |
|
|
|
68 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
69 |
|
|
|
70 |
|
|
|
|