A5005014399- Remote-Sensing Image Classification
- Advanced Image Fusion Techniques
- Remote Sensing and Land Use
- Geochemistry and Geologic Mapping
- Calibration and Measurement Techniques
- Satellite Image Processing and Photogrammetry
- Infrared Target Detection Methodologies
- Remote Sensing in Agriculture
- Planetary Science and Exploration
- Urban Heat Island Mitigation
- Remote Sensing and LiDAR Applications
- Advanced Image and Video Retrieval Techniques
- Spectroscopy and Chemometric Analyses
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Geophysics and Gravity Measurements
- Advanced Frequency and Time Standards
- Robotics and Sensor-Based Localization
- Astro and Planetary Science
- Cryospheric studies and observations
- Image and Signal Denoising Methods
- Soil Geostatistics and Mapping
- Spacecraft Design and Technology
- Inertial Sensor and Navigation
- Polar Research and Ecology
- Atmospheric and Environmental Gas Dynamics
The University of Tokyo
2014-2023
Japan Aerospace Exploration Agency
2008-2020
Vaughn College of Aeronautics and Technology
2019
Harbin Institute of Technology
2019
ETH Zurich
2019
Niigata University
2019
Mitsubishi Electric (Japan)
2019
Kyoto University
2019
Institute of Space and Astronautical Science
2008-2018
New Energy and Industrial Technology Development Organization
2010-2011
Coupled nonnegative matrix factorization (CNMF) unmixing is proposed for the fusion of low-spatial-resolution hyperspectral and high-spatial-resolution multispectral data to produce fused with high spatial spectral resolutions. Both are alternately unmixed into end member abundance matrices by CNMF algorithm based on a linear mixture model. Sensor observation models that relate two built initialization each NMF procedure. This physically straightforward easy implement owing its simple update...
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model (GDEM) was released to the public in My 2009. GDEM version 2, currently under development, is reproduced using an updated algorithm contains new ASTER data observed after September 2008. Validation study of trial beta 2 carried out investigate its characteristics such as geolocation error, elevation error horizontal resolution. results confirmed that offset resolution will be greatly...
The mechanism of crosstalk phenomena in the shortwave infrared (SWIR) subsystem Advanced Spaceborne Thermal Emission and Reflection Radiometer, which has six bands wavelength region 1.6-2.43 /spl mu/m, is investigated. It found that light incident to band 4 reflected at detector filter boundary, then transported other by multiple reflections focal plane area. A correction algorithm developed improve spectral separation performance SWIR. Parameters model, i.e., amount stray its area...
We discovered a vertical hole on the Moon, which is possible lava tube skylight, using data from SELENE's two high‐resolution cameras: Terrain Camera and Multi‐band Imager. The nearly circular, 65 m in diameter, located sinuous rille at Marius Hills region, volcanic province lunar nearside. observed various solar illumination conditions estimated its depth to be 80 88 m. depth/diameter ratio much larger than for typical impact craters. There are neither conspicuous deposits indicating...
Classification techniques for hyperspectral images based on random forest (RF) ensembles and extended multiextinction profiles (EMEPs) are proposed as a means of improving performance. To this end, five strategies - bagging, boosting, subspace, rotation-based, boosted rotation-based used to construct the RF ensembles. EPs, which an extrema-oriented connected filtering technique, applied associated with first informative components extracted by independent component analysis, leading set...
The classification of tree species in forests is an important task for forest maintenance and management. With the increase spatial resolution remote sensing imagery, individual next target research area inventory. In this work, we propose a methodology involving combination hyperspectral LiDAR data classification, which can be extended to areas shadow caused by illumination crowns with sunlight. To remove influence shadows data, unmixing-based correction applied as preprocessing. Spectral...
At the core of technical methodology for creating Advanced Spaceborne Thermal Emission and Reflection radiometer (ASTER) global digital elevation model (GDEM) is procedure generating a set 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">°</sup> latitude-by-1 longitude tiles containing DEM data in geographic latitude coordinates with one arc second postings from scene-based ASTER DEMs. The GDEM comprised all tiles, which include at least 0.01%...
Nonlinear spectral mixture models have recently received particular attention in hyperspectral image processing. In this paper, we present a novel optimization method of nonlinear unmixing based on generalized bilinear model (GBM), which considers the second-order scattering photons model. Semi-nonnegative matrix factorization (semi-NMF) is used for to process whole form. When endmember spectra are given, abundance and interaction fractions converge local optimum by alternating update rules...
We present a novel method for detecting instances of an object class or specific in high-spatial-resolution optical remote sensing images. The proposed integrates sparse representations local-feature detection into generalized-Hough-transform detection. Object parts are detected via class-specific image patches using learned target and background dictionaries, their co-occurrence is spatially integrated by Hough voting, which enables aim to efficiently detect objects small set positive...
Hyperspectral Imager Suite (HISUI) is a next-generation Japanese optical sensor that composed of hyperspectral imager and multispectral imager, which will be launched on Advanced Land Observation Satellite 3 (ALOS-3). Combination imagers contributes land remote sensing with high spectral spatial resolution. Calibration the HISUI functional-equivalent model has been carried out to meet science requirements sensor. A center wavelength response each band obtained, from distortions in...
Multiple classifier systems or ensemble learning is an effective tool for providing accurate classification results of hyperspectral remote sensing images. Two well-known classifiers data are random forest (RF) and rotation (RoF). In this paper, we proposed to use a novel decision tree (DT) method, namely, canonical correlation (CCF). More specifically, several individual trees (CCTs) that binary DTs, which components the hyperplane splitting, used construct CCF. Additionally, adopt...
Accurate attitude estimation of spacecrafts is the main requirement to provide good geometric performance remote-sensing imagery. The Advanced Spaceborne Thermal Emission and Reflection Radiometer/short-wave-infrared subsystem has six linear-array sensors arranged in parallel, each line scans same ground target with a time interval 356.238 ms between neighboring bands. registration bands becomes worse when fluctuation occurs during lag observations. Since resolution scan higher than that...
The inside of Shackleton Crater at the lunar south pole is permanently shadowed; it has been inferred to hold water-ice deposits. Terrain Camera (TC), a 10-meter-resolution stereo camera onboard Selenological and Engineering Explorer (SELENE) spacecraft, succeeded in imaging crater, which was faintly lit by sunlight scattered from upper inner wall near rim. estimated temperature crater floor, based on shape model derived TC data, less than â¼90 kelvin, cold enough water-ice. However, TC's...
The data fusion of low spatial-resolution hyperspectral and high multispectral images enables the production with small spectral distortion. EO-1/Hyperion is world's first sensor. It was launched in 2001 has a similar orbit to Terra/ASTER. In this work, we apply Terra/ASTER datasets by preprocessing onboard cross-calibration sensor characteristics. relationship response function determined convex optimization comparing over same range. After accurate image registration, point spread obtained...
Due to the development of sensors and data acquisition technology, fusion features from multiple is a very hot topic. In this letter, use morphological fuse hyperspectral (HS) image light detection ranging (LiDAR)-derived digital surface model (DSM) exploited via an ensemble classifier. each iteration, we first apply openings closings with partial reconstruction on few principal components (PCs) HS LiDAR sets produce spatial elevation information for sets. Second, three groups (i.e.,...
Deep learning-based approaches for land cover segmentation rely on supervision with pixel-level ground truth, but may not generalize well to unseen image domains. Due the tedious and labor-intensive labeling process, transferring models trained label-rich source data nonannotated target becomes a popular problem in recent years. Because of domain shift, difference between distributions can degrade accuracy if training occurs directly without proper adaptation (DA). In this letter, we propose...
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) system acquires multispectral images ranging from the visible to thermal infrared region. ASTER consists of three subsystems: near-infrared (VNIR), short-wave (SWIR) (TIR) radiometers. VNIR subsystem has a backward-viewing telescope as well nadir one. To deliver data products high quality viewpoint geolocation band-to-band registration performance, fundamental program, called Level-1 processing, been developed for...
Image matching and water-body-detection methodologies are essential parts of generating good-quality digital elevation model (DEM) data. It is one the very important results for image where 1-D searching in along-track direction sufficient to find maximum correlation point if reconstructed unprocessed Advanced Spaceborne Thermal Emission Reflection data (Level-1A data) used as source DEM products. This situation obtained from general formulation how make possible. The quality evaluated this...
Coupled non-negative matrix factorization (CNMF) is introduced for hyperspectral and multispectral data fusion. The CNMF fused have little spectral distortion while enhancing spatial resolution of all band images owing to its unmixing based algorithm. applied the synthetic dataset generated from real airborne taken over pasture area. quality evaluated by classification accuracy types. experiment result shows that enables accurate identification observed materials at fine resolution.
Hyperspectral Imager Suite (HISUI) is a future spaceborne hyperspectral Earth imaging system being developed by Japanese Ministry of Economy, Trade, and Industry (METI). HISUI will be launched deployed on International Space Station (ISS) for three year operation from 2019. In FY2016, manufacturing testing Flight Model design Exposed Payload System were conducted. data policy research announcement as well synergy with other earth observing instruments onboard ISS are also discussed.