Open AccessComputer Vision Methods for Automotive Applications
Author(s) -
Aldo André Díaz Salazar,
Paulo Roberto Gardel Kurka
Publication year - 2020
Publication title -
tecnia
Language(s) - English
Resource type - Journals
eISSN - 2309-0413
pISSN - 0375-7765
DOI - 10.21754/tecnia.v30i2.801
Subject(s) - automotive industry , computer vision , computer science , artificial intelligence , camera resectioning , stereopsis , calibration , machine vision , stereo cameras , ellipse , set (abstract data type) , engineering , mathematics , programming language , aerospace engineering , statistics , geometry
Recent advances in computer vision have favored technological developments in the automotive industry. In this work, we present methods relevant to the use of cameras as measurement devices in computer vision and their applications in the automotive industry. The methods are edge and ellipse detection, camera calibration, 3-D reconstruction and stereo vision. The applications include methods for detecting rims in automotive wheels, estimation of the calibration angles of vehicles and the reconstruction of a vehicle's trajectory using stereo vision. The results showed the potential of computer vision methods to solve complex problems in the automotive industry. In conclusion, a set of techniques and applications of computer vision in the automotive industry are presented in order to motivate future developments in this and other related areas.