Desarrollo e implementación de un prototipo de parqueadero inteligente a través de técnicas de visión artificial

Abstract

Abstract: The present work is focused on the detection of vehicles in a parking lot in real time through a surveillance camera, this implies that the system must know at each moment where the object to be detected is located, in turn there are many factors that can affect detection, such as occlusions, shadows, lighting, etc. Due to this, we will make use of different techniques that help the correct functioning of object detection. According to the investigations, the use of deep learning algorithms and neural networks is proposed, capable of reducing the external factors that produce a poor detection of vehicles and malfunction of the system in general, as well as the use of various image processing techniques applied to the vehicle detection experimentation and analysis. Finally, the research project allows us to know which techniques and algorithms are the best to use in the detection of vehicles, carrying out experiments in real time and with pre-recorded videos of parking lots, in this way it will be possible to know the best alternative in the implementation of a vehicle detector.