講演詳細

経営部門[Forest Management]

日付	2024年3月8日
開始時刻	ポスター発表
会場名	531
講演番号	PD-36（学生ポスター賞審査対象）
発表題目	Land Use and Land Cover Classification of Mangrove Area in Myanmar Using Deep Learning and Remote Sensing Dataset Land Use and Land Cover Classification of Mangrove Area in Myanmar Using Deep Learning and Remote Sensing Dataset
所属	The University of Tokyo
要旨本文	Accurate information on land use and land cover (LULC) is crucial for effective regional land and forest management. This study addresses the challenge of obtaining reliable LULC information in an intricate mangrove ecosystem by employing advanced deep learning models with multisource satellite images. U-Net and artificial neural network models were trained and tested using labeled images created from field ground truth points and evaluated for each class. The results demonstrated that the combination of PlanetScope bands, spectral indices, DEM, and CHM yielded superior performance for both models compared to the Sentinel-2 dataset. The proposed classification method produced a reliable and up-to-date LULC map of the Wunbaik Mangrove Area in Myanmar, providing practical implications for implementing conservation measures.
著者氏名	○Win, Sithu Maung ・ Satoshi Tsuyuki
著者所属	Department of Global Agricultural Sciences, Graduate School of Agriculture and Life Sciences, The University of Tokyo
キーワード	Land Use and Land Cover Classification, Deep Learning, Remote Sensing, Mangrove
Key word	Land Use and Land Cover Classification, Deep Learning, Remote Sensing, Mangrove