Computer Vision Based Sensor System for Autonomous Vehicles

Transport Means 2021: Sustainability: Research and Solutions: Proceedings of the 25th International Scientific Conference. Part 3 2021
Mihails Gorobecs, Ambuja Bangalore Srinivasa

This study is devoted to solve the perception problem of autonomous transport system by Convolutional Neural Network (CNN) application. The goal of this paper is to develop a functional software based sensor to detect and recognize objects in real-time, as well as estimate the speed of detected vehicle using solo camera as sensor. This paper aims at evaluating the workability of this sensor in different weather conditions for autonomous vehicle applications. This sensor can be applied for different levels of abstraction like road safety, autonomous drone navigation, guiding system for disabled, surveillance to name few.

Atslēgas vārdi
autonomous vehicles, CNN, object recognition, speed estimation, computer vision
Hipersaite
https://transportmeans.ktu.edu/wp-content/uploads/sites/307/2018/02/Transport-Means-2021-Part-III.pdf#page=207

Gorobecs, M., Bangalore Srinivasa, A. Computer Vision Based Sensor System for Autonomous Vehicles. No: Transport Means 2021: Sustainability: Research and Solutions: Proceedings of the 25th International Scientific Conference. Part 3, Lietuva, Kaunas, 6.-8. oktobris, 2021. Kaunas: Kaunas University of Technology, 2021, 1175.-1180.lpp. ISSN 1822-296X. e-ISSN 2351-7034.

Publikācijas valoda
English (en)