Data in Brief · February 26, 2026
Kazakh banknote image dataset
Ualikhan Sadyk, Makhambet Yerzhan, Cemil Turan, Haohan Wang
Abstract
This paper presents an image dataset of Kazakh banknotes collected for research in computer vision and pattern recognition. The dataset contains photographs of banknotes from seven categories corresponding to the denominations 500, 1000, 2000, 5000, 10,000, 20,000 Kazakhstani tenge, as well as a mixed category containing multiple denominations within a single image. Each denomination folder includes at least 100 images, while the mixed category contains over 1000 images. Images were captured against a wide range of backgrounds and under varying lighting conditions to reflect realistic usage scenarios and environmental diversity. The 5000 tenge denomination is further divided into three subsets that distinguish between old and new banknote designs and between single-note and multiple-note configurations. Most images contain multiple banknotes of the same denomination, except for one subset explicitly designed to include a single banknote per image. All files follow a consistent numeric naming convention. The dataset is intended to support tasks such as banknote classification, object detection, and robustness analysis of deep learning models under background and illumination variation. It may also be reused for transfer learning, dataset bias studies, and comparative evaluations across currency recognition systems.
More by Ualikhan Sadyk
- Interpretable Hybrid CNN-BiLSTM-Attention Model for ECG Arrhythmia ClassificationIEEE International Conference on Smart Information Systems and Technologies · 2026
- Development of Novel Neural Network Architecture for Automated Pharmaceutical and Vitamin Identification Using Image ClassificationInternational Conference on Electronics, Computer and Computation · 2026
- Development of Multimodal Neural Network for Suicidal Ideation Detection Using Kazakh-Russian BERT and Attention MechanismInternational Conference on Electronics, Computer and Computation · 2026
- Benchmarking Deep Transfer Learning for Banknote Recognition: Vision Transformers Vs. CNNs on the KZ-BD DatasetInternational Conference on Electronics, Computer and Computation · 2026