A dataset of road traffic images taken from unmanned aerial vehicles (UAV) with the purpose of being used to train artificial vision algorithms, among which those based on convolutional neural networks stand out. Dataset is available and accessible in order to improve the performance of road traffic vision and management systems due to the lack of resources in this specific domain. The full description of the characteristics of the dataset, as well as its components and format, can be found here: https://www.mdpi.com/2306-5729/7/5/53 The dataset is also available on Kaggle: https://www.kaggle.com/datasets/javiersanchezsoriano/traffic-images-captured-from-uavs

The dataset format is YOLO. Relevant data about the dataset: Scenes Frames Targets Cars Motorbikes Regional road 4.500 24.858 14.577 10.281 Urban intersection 2.462 10.759 10.759 0 Rural road 1.292 746 746 0 Split roundabout 2.297 3.107 3.107 0 Roundabout (Far) 1.814 71.819 64.844 6.975 Roundabout (Near) 3.997 4.4039 43.569 470 Total 15.070 155.328 137.602 17.726

If you use this dataset please cite this paper: Bemposta Rosende, S.; Ghisler, S.; Fernández-Andrés, J.; Sánchez-Soriano, J. Dataset: Traffic Images Captured from UAVs for Use in Training Machine Vision Algorithms for Traffic Management. Data 2022, 7, 53. https://doi.org/10.3390/data7050053