Detecção De Patologias Em Rodovias Utilizando Métodos De Aprendizado Profundo

Abstract

This research aims to develop an effective system for detecting road damage on highways using computer vision techniques, thereby enhancing drivers' safety and improving road maintenance. Computer vision, an interdisciplinary field that encompasses image processing and pattern recognition, has applications far beyond highway damage detection, extending into various other domains. Over the years, several methods have been proposed for road damage detection, including traditional image processing, deep learning, and stereo vision. In this study, the YOLO (You Only Look Once) algorithm was employed to detect damage on road pavements. The research objectives include fostering scientific research skills, providing an integrated perspective on different topics, and contributing to the reduction of accidents and the maintenance costs associated with highways.