Automatic recognition of dangerous objects in front of the vehicle's windshield

In this article, an automatic recognition system, of objects that threaten the safety of vehicle users is proposed. These objects are detected in front of the vehicle's windshield based on images from video recorders and then classified. The tests were conducted using artificial neural networks...

Full description

Saved in:
Bibliographic Details
Published inSignal Processing Algorithms, Architectures, Arrangements, and Applications Conference proceedings pp. 91 - 96
Main Authors Balcerek, Julian, Pawlowski, Pawel, Cierpiszewska, Julia, Kus, Lukasz, Zborowski, Piotr
Format Conference Proceeding
LanguageEnglish
Published Division of Signal Processing and Electronic Systems, Poznan University of Technology (DSPES PUT) 25.09.2024
Subjects
Artificial neural networks
Automobiles
Automotive components
Birds
Ice
image processing
image recognition
Recording
Safety
Signal processing algorithms
threat object recognition
video recorders
vision system
Wheels
Online AccessGet full text
ISSN2326-0319
DOI10.23919/SPA61993.2024.10715613

Cover

More Information
Summary:In this article, an automatic recognition system, of objects that threaten the safety of vehicle users is proposed. These objects are detected in front of the vehicle's windshield based on images from video recorders and then classified. The tests were conducted using artificial neural networks on a specially prepared database of frames from real recordings. Recognized objects include: car wheels, car parts, birds, stones and ice. They were recognized in single- and multi-class detections for various parameters of the network model. Detection efficiency of 95 \% was achieved in the best case.
ISSN:2326-0319
DOI:10.23919/SPA61993.2024.10715613