Dataset for Detecting Vehicle in Pedestrian Areas version 2

The Car_Truck_Bus_Pedestrian-Area_v2.zip is an updated collection featuring a greater number of images, specifically 7808 images, each in very high resolution in .jpg format. The total file size is now 290 MB. It continues to serve the purpose of detecting vehicles in pedestrian areas within urban environments and is captured at a resolution of 1280×720 (HD). Each image is approximately 58 KB in size. The image data was collected from 100 strategically placed public CCTVs in Medellín, Colombia, under various light and weather conditions, recorded every 60 seconds between 8:00 a.m. and 3:00 p.m. For each image, it is necessary to identify cars, buses, trucks, and pedestrians in particular zones, differentiated from the background. This enhanced dataset aims to train and evaluate machine learning models for urban traffic management and pedestrian safety applications. Data in this document is published under the Creative Commons Attribution 4.0 International License.