Semantic segmentation using support vector machine classifier

0571888 - ÚGN 2024 RIV CZ eng C - Conference Paper (international conference)
Pecha, Marek - Langford, Z. - Horák, David - Tran Mills, R.
Semantic segmentation using support vector machine classifier.
Programs and Algorithms of Numerical Mathematics 21 : Proceedings of Seminar. Praha: Institute of Mathematics CAS Prague, 2023 - (Chleboun, J.; Kůs, P.; Papež, J.; Rozložník, M.; Segeth, K.; Šístek, J.), s. 173-186. ISBN 978-80-85823-73-8.
[Programs and Algorithms of Numerical Mathematics /21./. Jablonec nad Nisou (CZ), 19.06.2022-24.06.2022]
EU Projects: European Commission(XE) 847593 - EURAD
Institutional support: RVO:68145535
Keywords : wildfire identification * semantic segmentation * support vector machines * distributed training
OECD category: Applied mathematics
https://dml.cz/bitstream/handle/10338.dmlcz/703198/PANM_21-2022-1_19.pdf

This paper deals with wildfire identification in the Alaska regions as a semantic segmentation task using support vector machine classifiers. Instead of colour information represented by means of BGR channels, we proceed with a normalized reflectance over 152 days so that such time series is assigned to each pixel. We compare models associated with $\mathcal{l}1$-loss and $\mathcal{l}2$-loss functions and stopping criteria based on a projected gradient and duality gap in the presented benchmarks.
Permanent Link: https://hdl.handle.net/11104/0342781