A dataset of video stream at traffic intersections in valley-city Lanzhou under complex scenes

<p indent="0mm">Lanzhou, a typical valley city, faces great challenges in traffic management due to its unique geographical environment and complex traffic layout. To address this challenge, this study has constructed a dataset of video stream at traffic intersections in valley-city Lanzhou under complex scenes. The dataset covers video surveillance data from several key traffic intersections in Lanzhou City, covering different time periods (such as morning peak and off-peak hours), weather conditions (such as sunny, foggy, etc.), traffic conditions (such as congestion, smooth flow, etc.), and lighting conditions (such as dim light, full light, etc.). The data was collected on November 16, 2021, December 3, 2021, and December 9, 2021. In terms of data processing, this study has meticulously labeled and classified the video data, such as vehicle type, vehicle lane and so on. The dataset in this study has undergone rigorous quality control procedures, including integrity verification, clarity evaluation, stability check, and annotation accuracy verification, ensuring the high reliability of the dataset. This dataset is rich and diverse, aiming to provide solid data support for the development of intelligent transportation systems, traffic congestion control, and traffic accident prevention, and thereby providing an empirical basis for traffic flow analysis in river valley cities.