Ugandan banknote image dataset for genuine and counterfeit classification

This dataset contains labelled images of Ugandan banknotes prepared for genuine and counterfeit classification tasks. The images cover common Ugandan shilling denominations, including UGX 1000, 2000, 5000, 10000, 20000, and 50000, and were collected from mobile money agent shops, micro-finance institutions, banks, and local markets in Uganda. The dataset is organized into two classes: genuine and counterfeit. Images were captured using smartphone cameras under everyday field conditions to reflect realistic banknote-handling environments. The cleaned dataset includes deduplicated images, standardized filenames, and train/validation/test splits. The dataset can support research and teaching in computer vision, image classification, counterfeit detection, currency authentication, and applied machine learning for financial security.