BatSilver/TextDetectionData

Text Detection Dataset

This dataset is a comprehensive collection of Arabic and English text detection samples designed for benchmarking and evaluating text detection models. The dataset combines samples from multiple open-source datasets to provide diverse text detection challenges across different domains, scripts, and image conditions.

---

📁 Dataset Structure

• Images are stored in .jpg format in the "Text Detection Images" folder
• Accompanied by a CSV file (textdetectionannotations.csv) with bounding box coordinates and image characteristics.

---

📚 Data Sources and Attribution

This dataset is compiled from samples taken from the following sources:

• SROIE Dataset v2: Scene text detection samples from Kaggle - urbikn
• ICDAR2013: International Conference on Document Analysis and Recognition 2013 dataset samples
• EvArEST Dataset: Arabic scene text samples from HGamal11/EvArEST-dataset-for-Arabic-scene-text
• Arabic Documents OCR Dataset: Document text detection samples from Kaggle - humansintheloop

---

📊Dataset Statistics

Metric	Value
Total Images	725
Languages	English and Arabic
Image Format	.jpg
Annotation Type	Bounding Box Coordinates in Polygon format
Source Datasets	4 (SROIE, ICDAR2013, EvArEST, Arabic Documents OCR)

---

📊 Data Insights

🌍Dataset Characteristics

Characteristic	Values
Dataset Name	SROIE, ICDAR2013, EvArEST, Arabic Documents OCR
Image Category	receipt, scene_text
Language	Arabic, English
Image Quality
- Blurry	True, False
- Noisy	True, False
- Rotated	True, False
Contrast Level	low, medium, high
Brightness Level	low, medium, high
Resolution Category	low, medium, high
Complexity Level	low, medium, high

---

🛠️ How to Use

You can load the dataset using:

from datasets import load_dataset

ds = load_dataset("BatSilver/TextDetectionData")