A4 Refereed article in a conference publication
Study on Stereo AI Based Zed-2i Camera
Authors: Padberg, Timon; Heikkonen, Jukka; Kanth, Rajeev
Editors: Rocha, Álvaro; Ferrás, Carlos; Hochstetter Diez, Jorge; Diéguez Rebolledo, Mauricio
Conference name: Information Technology and Systems
Publisher: Springer Nature Switzerland
Publication year: 2024
Journal: Lecture Notes in Networks and Systems
Book title : Information Technology and Systems : ICITS 2024, Volume 2
Volume: 933
First page : 46
Last page: 56
ISBN: 978-3-031-54255-8
eISBN: 978-3-031-54256-5
ISSN: 2367-3370
eISSN: 2367-3389
DOI: https://doi.org/10.1007/978-3-031-54256-5_5
Publication's open availability at the time of reporting: No Open Access
Publication channel's open availability : Partially Open Access publication channel
Web address : https://doi.org/10.1007/978-3-031-54256-5_5
This article examines the possibility of successfully integrating the ZED-2i stereo camera with AI-enhanced technology. A sample of two different environments with different characteristics is assessed to determine the viability of implementing this in a real-world setting. The results show a significant difference in cloud detection between indoor and outdoor environments, with higher detail levels in point clouds. Essential factors include depth recognition, cloud clustering, and light conditions. Quality parameter settings were found to mitigate the adverse effects of localization and visual odometry inaccuracies. The findings from analyzing experiments display the importance of hardware and software configurations. Future research is required on the newer version of the camera with Jetson NX Orin implementation to explore and develop the utilization in specific sectors like agriculture and forestry.
