Abstract In autonomous driving, the ability to correctly detect and calculate the distance to objects is a fundamental task of the perception system. The objective of this paper is to present a robust object detection system using a Time of Flight (ToF) camera. The system is designed for covering the blind spot areas of vehicles and to be integrated with an autonomous vehicle perception system. ToF cameras generate images with depth information in real time as well as providing grayscale images. The accuracy of the measurements obtained depends greatly on the correct calibration of the camera, therefore in this paper, a calibration process for the Sentis 3D-M420 ToF camera is presented. For the object detection system that has been developed, different descriptors and classifiers have been analyzed. To evaluate the system tests were performed in real situations using a series of images obtained with the Sentis 3D-M420 camera during autonomous driving tests.
Time of Flight Camera Calibration and Obstacle Detection Application for an Autonomous Vehicle
2019-09-05
10 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Obstacle Detection using a Time of Flight Range Camera
IEEE | 2018
|Camera Calibration for Lane and Obstacle Detection
British Library Conference Proceedings | 1999
|Autonomous vehicle obstacle detection and tracking
Tema Archiv | 1998
|Obstacle detection system for autonomous vehicle or robot
Europäisches Patentamt | 2023