CaptainNemo18/IndianCurrencyDataset

Indian Currency Recognition Dataset

Overview

The Indian Currency Recognition Dataset is a publicly available dataset designed for researchers and developers working on currency classification, currency recognition, and counterfeit currency detection using computer vision and deep learning.

The dataset contains high-resolution images of multiple Indian currency denominations captured under real-world conditions. It includes both old and new currency notes and has been augmented to improve model generalization.

---

Dataset Information

Attribute	Details
Dataset Name	Indian Currency Dataset
Version	1
Published	28 August 2020
DOI	10.17632/48ympv8jjf.1
License	CC BY 4.0
Total Original Images	4,657
Total Augmented Images	11,657
Image Format	JPEG
Camera Used	Moto X Play (21 MP)

---

Supported Currency Denominations

The dataset contains images of the following Indian currency notes:

• ₹10 (Old & New)
• ₹20 (Old)
• ₹50 (Old & New)
• ₹100 (Old & New)
• ₹200
• ₹500
• ₹2000

Dataset folders:

Rs.10/
Rs.20/
Rs.50/
Rs.100/
Rs.200/
Rs.500/
Rs.2000/

---

Image Specifications

Landscape Images

• Resolution: 5344 × 3006

Portrait Images

• Resolution: 3006 × 5344

All images were captured using a Moto X Play smartphone equipped with a 21 MP camera.

---

Data Augmentation

To increase dataset diversity and improve deep learning performance, the following augmentation techniques were applied:

• Zoom
• Rotation (90°)
• Rotation (270°)
• Tilt
• Distortion
• Horizontal Flip

After augmentation, the dataset size increased from 4,657 images to 11,657 images.

---

Applications

This dataset can be used for:

• Indian Currency Recognition
• Currency Classification
• Deep Learning Research
• Computer Vision Projects
• Assistive Technology for Visually Impaired Individuals
• Counterfeit Currency Detection
• Mobile Currency Recognition Applications

---

Dataset Structure

Indian-Currency-Dataset/
│
├── Rs.10/
├── Rs.20/
├── Rs.50/
├── Rs.100/
├── Rs.200/
├── Rs.500/
├── Rs.2000/
└── README.md

---

Source

This dataset was developed as part of the project:

"Coinnet: Platform-independent Application to Recognize Indian Currency Notes Using Deep Learning Techniques."

The work was carried out during a sabbatical at Bennett University with support from SR University (formerly SR Engineering College), Warangal.

---

Citation

If you use this dataset in your research or project, please cite:

@dataset{veeramsetty2020indiancurrency,
  author = {Venkataramana Veeramsetty and Gaurav Singal and Tapas Badal},
  title = {Indian Currency Dataset},
  year = {2020},
  publisher = {Mendeley Data},
  version = {V1},
  doi = {10.17632/48ympv8jjf.1}
}

Citation:

Veeramsetty, V., Singal, G., & Badal, T. (2020). Indian Currency Dataset (Version 1). Mendeley Data. https://doi.org/10.17632/48ympv8jjf.1

---

Related Resources

• GitHub Repository: https://github.com/10zinten/Indian-Currency-Recognition
• Research Paper: https://link.springer.com/article/10.1007/s11042-020-09031-0

---

License

This dataset is distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License.

You are free to:

• Share
• Adapt
• Build upon the dataset

provided appropriate credit is given to the original authors.

For more information, visit:

https://creativecommons.org/licenses/by/4.0/

---

Acknowledgements

Special thanks to:

• Venkataramana Veeramsetty
• Gaurav Singal
• Tapas Badal

for collecting and publishing this dataset for the research community.