Determining the Occupancy of Vehicle Parking Areas by Deep Learning

摘要

Parking a vehicle in heavy traffic situations leads to prolonged driving time, deterioration of traffic flow and therefore environmental pollution when searching for free space. Although the sensor systems in the indoor parking lots are beneficial, these systems cannot be applied to outdoor spaces. In this study, a deep learning application was developed which classifies the occupancy status of the parking spaces in outdoor parking areas. High accuracy rates were obtained in this application where transfer learning was performed using ResNet model.