Abstract: In the modern era, the problem of parking is also growing because of the growth in the number of vehicles. From the closing decade, numerous researches took place intending to broaden a perfect automated parking slot occupancy detection. There is an auto mechanism that can park vehicles automatically but it is required to detect which parking slot is available and which one is busy. It is proposed to design and implement a parking space detection using image processing. Images are captured when a car enters or leaves the parking slot. Computer vision techniques are used to infer the state of the parking slot given the data collected from the database. This project presents an approach for a real-time parking space classification based on Convolutional Neural Networks (CNN) using the Mask R-CNN framework. The training process has been done using MR-CNN and the output is a pickled model used for predictions to detect vacant and occupied parking slots. The system checks a defined area whether a parking slot (bounding boxes defined at initialization of the system) is containing a car or not (occupied or vacant). The proposed system indicates the great potential of this method to provide a low-cost and reliable solution to the PG (Parking Guidance) systems in outdoor environments.
Keywords: Convolution Neural Network, Parking Guidance
Copyright Form