Dataset Of Kazakhstan Banknotes with Annotations

Recognizing and classifying currency is a significant challenge for individuals with visual impairments. It can greatly enhance their independence when it comes to financial transactions. To address this, we've curated a comprehensive dataset of Kazakhstani banknotes, complete with annotations. This dataset comprises a total of 4200 high-quality images of Kazakhstani banknotes. For our Kazakhstani banknote dataset, we've categorized it into 14 classes to cover the full range of denominations. These classes include 1 KZT, 2 KZT, 5 KZT, 10 KZT, 20 KZT, 50 KZT, 100 KZT, 200 KZT, 500 KZT, 1000 KZT, 2000 KZT, 5000 KZT, 10000 KZT, and 20000 KZT. Each class represents a different denomination of Kazakhstani currency. The dataset features images captured using the rear cameras of mobile phones, taken under various conditions. This includes different lighting environments, cluttered backgrounds, and even folded banknotes to simulate real-world scenarios. We've also meticulously annotated the images in the YOLO (You Only Look Once) format. The dataset is divided into "Training" and "Validation" subsets, maintaining an 70:30 ratio to ensure robust model training and evaluation. Our aim is to empower individuals with visual impairments by providing them with the tools they need for seamless and independent financial transactions.