Object Detection Through Computer Vision

This research work is focused on the field of object detection through computer vision, focusing on a very popular topic nowadays, smart cities. Therefore, a proposal is presented, after evaluating the different systems that have been used in the literature, based on deep learning techniques together with the use of images to detect traffic density. The developed technique corresponds to a You Only Look Once (YOLO) algorithm, which allows using previously trained weights or training the weights with a personalized dataset, for traffic detection. In relation to the results obtained after the application of these techniques, it can be observed how, depending on the data set used for training, the algorithm obtains different detections.