Generation of the High-Resolution Land-Use and Land-Cover Map in Japan Version 21.11
Sota Hirayama, Takeo Tadono, Yousei Mizukami, Masato Ohki, Koichi Imamura, Naoyoshi Hirade, Fumi Ohgushi, Masanori Dotsu, Tsutomu Yamanokuchi, Kenlo Nishida Nasahara
Abstract
Japan Aerospace Exploration Agency (JAXA) Earth Observation Research Center (EORC) has produced a high-resolution land-use and land-cover classification map for Japan indicating the average cover in 2018–2020. The classification accuracy was improved by using a combination of Sentinel-2 and ALOS-2/PALSAR-2 data. We developed a convolutional neural network-based algorithm specialized for pattern recognition in multispectral and time-series feature spaces, and achieved an overall accuracy of 88.85% in 12 classifications. This is the most accurate high-resolution land-cover map ever produced by JAXA. This highly accurate product is being used for the calculation of SDGs global indicators such as 15.4.2 “Mountain Green Cover Index”.