Yemeni Currency Recognition Dataset

This dataset contains real-world images of Yemeni banknotes collected for research in currency recognition and computer vision. The images were captured using smartphone cameras under diverse conditions, including variations in illumination, background, orientation, scale, and partial occlusions.

The dataset includes six Yemeni Riyal denominations: 50, 100, 200 (old), 250, 500, and 1000 YER. Each denomination is organized in a separate folder. The dataset consists of a total of 1,691 RGB images.

This dataset is intended to support tasks such as image classification, object detection, and evaluation of currency recognition systems, including assistive applications for visually impaired users. It is designed to facilitate reproducible research and testing of deep learning models under realistic acquisition conditions.

本数据集收录了用于货币识别与计算机视觉研究的也门纸币实拍图像。上述图像均由智能手机摄像头在多样化采集环境下拍摄，涵盖光照、背景、拍摄角度、图像尺度及局部遮挡等多种变化场景。

本数据集包含六种也门里亚尔（YER）面额：50、100、200（旧版）、250、500及1000也门里亚尔。各类面额的图像均存储于独立文件夹中，数据集总计包含1691张RGB图像。

本数据集旨在支撑图像分类、目标检测以及货币识别系统（包括面向视障用户的辅助应用）的评估等相关任务，其设计初衷在于助力可复现研究，并便于在真实采集条件下对深度学习模型开展测试与验证。