Patch size selection for analysis of sub-meter resolution hyperspectral imagery of forests

Matti Mõttus, Matthieu Molinier, Eelis Halme, Hai Cu, Jorma Laaksonen

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

Abstract

Very high resolution remote sensing data of forests, where individual tree crowns are separable, contains structural information on tree size and density. Such information is complementary to the spectral signatures currently used in forestry applications. Advanced machine learning methods, e.g. convolutional neural networks (CNNs), offer an automated and standardized way of retrieving both spectral and structural information from imagery. A key characteristic in CNNs is patch size, which should be large enough to include dominant structural scale, yet as small as possible to avoid unnecessary averaging. Our results show that the patch should be larger than one tree, but increasing it excessively reduces retrieval accuracy. Furthermore, large patch sizes can cause loss of independence between training and validation data, leading to overestimating model performance.
Original languageEnglish
Title of host publication2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS
PublisherIEEE Institute of Electrical and Electronic Engineers
Pages9007-9010
ISBN (Electronic)978-1-6654-0369-6, 978-1-6654-0368-9
ISBN (Print)978-1-6654-4762-1
DOIs
Publication statusPublished - 2021
MoE publication typeA4 Article in a conference publication
EventIEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 - Brussels, Belgium, Brussels, Belgium
Duration: 11 Jul 202116 Jul 2021

Conference

ConferenceIEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021
Country/TerritoryBelgium
CityBrussels
Period11/07/2116/07/21

Fingerprint

Dive into the research topics of 'Patch size selection for analysis of sub-meter resolution hyperspectral imagery of forests'. Together they form a unique fingerprint.

Cite this