Image Processing Toolbox

• Autor(a) principal: Castro, Diogo Filipe Miranda Ferreira
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Texto Completo: http://hdl.handle.net/10362/167930
• topic: Image Processing; Computer Vision; Digital Image; Digital Filter; Image Processing Library; Python; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Huge amounts of information regarding image processing and computer vision exist in the form of articles, books and courses. Most of them are too dense to understand or out of reach for most people. Nowadays most applications and scientific papers require certain advanced knowledge to understand its contents and make use of them. This leads for the need to recap the principles of image processing and computer vision, creating a path to the more complex and relevant concepts of today. In this thesis a structured document is developed explaining the principles and most common algorithms and methodologies, alongside with practical examples for each of them, attempting to solve this issue. In addition, this thesis’ work includes the creation of a new open-source library dedicated to understand and learn how each algorithm and method is implemented and what happens to each image pixel during manipulation. The results present in this work relate to the comparison of the implemented methods through three metrics, execution time, image quality and features extracted. It is discussed the performance of these metrics for each method. The basics of image processing and computer vision can be learned from this work, as well as some algorithms and concepts from the image processing and computer vision communities.