Home // SIGNAL 2021, The Sixth International Conference on Advances in Signal, Image and Video Processing // View article

Classifying Vessels in Inland Waters Using Live Video Streams

Authors:
Tomasz Hyla
Natalia Wawrzyniak

Keywords: classification; video surveillance; traffic monitoring; inland waters

Abstract:
Video surveillance systems used in vessels' traffic monitoring are being improved to eliminate the need for operator supervision. One desirable feature of such systems is the ability to count the number of passing vessels and determine their types. The paper presents straightforward algorithms that can determine the category of a vessel based on partial results of classification. The algorithms take as an input a series of vessel images acquired directly from the video stream by the detection module. Some of the images contain an inaccurately cropped ship or a part of it, and there are also occasions when ships or other objects in the background are erroneously cropped. The algorithms were implemented and experimental results from 100 video samples in four different qualities are presented. The results show that the method is suitable for a practical application.

Pages: 1 to 4

Copyright: Copyright (c) IARIA, 2021

Publication date: May 30, 2021

Published in: conference

ISSN: 2519-8432

ISBN: 978-1-61208-858-7

Location: Valencia, Spain

Dates: from May 30, 2021 to June 3, 2021