Kazakh Banknote Image Dataset

This dataset contains images of Kazakh banknotes across seven denominations: 500, 1000, 2000, 5000, 10000, 20000, and a mixed category (Mix) containing multiple denominations within the same image. Each denomination folder includes at least 100 images, with the exception of the Mix folder, which contains more than 1000 images. The banknotes are photographed against various backgrounds, providing diversity in lighting and scene conditions.

The 5000 KZT denomination is further divided into three subcategories:
- 5000_new_single — each image contains a single new banknote,
- 5000_new — images with multiple new-design banknotes,
- 5000_old — images with multiple old-design banknotes.

Most images contain multiple banknotes of the same denomination, except in the 5000_new_single subset. File names follow a numeric sequence (000001.jpg, 000002.jpg, …).

The dataset is designed for research in computer vision, including:
- banknote classification,
- object detection in images with multiple notes,
- studies on dataset diversity, transfer learning, and robustness of deep learning models in currency recognition tasks.

All images were collected and curated specifically for research purposes. The dataset does not contain any personal or sensitive information.