Browse by author
Lookup NU author(s): Dr Varun OjhaORCiD
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
© 2021, Springer Nature Switzerland AG. We investigate a deep transfer learning methodology to perform water segmentation and water level prediction on river camera images. Starting from pre-trained segmentation networks that provided state-of-the-art results on general purpose semantic image segmentation datasets ADE20k and COCO-stuff, we show that we can apply transfer learning methods for semantic water segmentation. Our transfer learning approach improves the current segmentation results of two water segmentation datasets available in the literature. We also investigate the usage of the water segmentation networks in combination with on-site ground surveys to automate the process of water level estimation on river camera images. Our methodology has the potential to impact the study and modelling of flood-related events.
Author(s): Vandaele R, Dance SL, Ojha V
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: 42nd DAGM German Conference on Pattern Recognition (DAGM GCPR 2020)
Year of Conference: 2021
Pages: 232-245
Print publication date: 17/03/2021
Online publication date: 16/03/2021
Acceptance date: 02/04/2018
ISSN: 0302-9743
Publisher: Springer
URL: https://doi.org/10.1007/978-3-030-71278-5_17
DOI: 10.1007/978-3-030-71278-5_17
Library holdings: Search Newcastle University Library for this item
Series Title: Lecture Notes in Computer Science
ISBN: 9783030712778