Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/88287
Title: OpenStreetCam: reconocimiento automático de objetos en imágenes mediante machine learning
Author: Villaluenga Morán, José Luis
Director: Merino, David  
Tutor: Muñoz Bollas, Anna
Abstract: OpenStreetCam is a project whose main objective is to map any geographical location using the images provided by its users. One of the most interesting features which provides the service, which has been the main object for the development of this work, is the detection in images of different elements such as traffic signs, people and vehicles. To identify the different entities listed, the system uses an open-source algorithm that applies the Machine Learning concept for the automatic detection of objects. In addition to the I.A (artificial intelligence) algorithm, the system uses a computer vision library developed by Facebook called RetinaNet. Among its features stands out its network model, which gives it superior performance to other existing detectors giving it a precision and speed that makes it suitable for use in real time applications. The objective of this work is to study the functioning of the OpenStretCam detaction system, as well as the evaluation of its performance and degree of resolution.
Keywords: algorithms
machine learning
OpenStreetCam
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: Jan-2019
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Treballs finals de carrera, treballs de recerca, etc.

Files in This Item:
File Description SizeFormat 
jvillaluengaTFG0119memoria.pdfMemoria del TFG2,62 MBAdobe PDFThumbnail
View/Open