Multidimensional high-resolution remote sensing based on spectral data fusion

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K20955
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent informatics; Perceptual information processing
• Research Institution: Institute of Physical and Chemical Research (2017); The University of Tokyo (2015-2016)
• Principal Investigator: Yokoya Naoto 国立研究開発法人理化学研究所, 革新知能統合研究センター, ユニットリーダー (40710728)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: リモートセンシング / スペクトル画像 / データ融合 / 物質・含有率分解 / 土地被覆変化認識 / 高解像度化

Outline of Final Research Achievements

We developed a data fusion technology to analyze multisensor optical images obtained from different Earth observation satellites to recognize the state and change of the ground surface with high resolution in spatial, temporal, and spectral domains. We developed a methodology that determines the materials and their fractional abundances at each pixel from multiple optical images obtained by different sensors. The proposed methodology is robust against the difference of sensor characteristics and observation conditions. We applied the technique to land-use and land-cover mapping, mineral mapping, and land-cover change detection, and demonstrated detailed identification of ground surface materials with high spatial and temporal resolution.

Free Research Field

空間情報学