Dataset for Real-Time Passenger Monitoring in Double-Decker Buses Using YOLOv8

This dataset supports the manuscript "Strategic Approaches to Real-Time Passenger Monitoring in Double-Decker Buses Using YOLO Technologies".It includes annotated training data, raw images, an external test dataset, and all machine-readable source data used to generate the tables and figures in the paper.The dataset is organized as follows:- Raw_Training_Dataset_412: Raw images used for training (Set A, N = 412)- Annotated_Dataset_412: Corresponding annotated dataset with rectangular bounding boxes- External_Dataset_80: Independent test dataset (N = 80), collected from 2025-06-16 to 2025-07-14- Table1–Table10.csv: Source data corresponding to all tables in the manuscript- Figure*_Source_Data.csv: Source data for figures, including training curves and distributions- Table3,4_External_Test_Result.csv: Per-image predictions on the external test set (true count, predicted count, and error)- SetA_0.1.csv – SetA_1.csv: Results from learning curve experiments using 10% to 100% of the training datasetAll data are provided in machine-readable CSV format to enable full reproducibility of the reported results.