51 Hits in 6.3 sec

Interpretable Hyperspectral AI: When Non-Convex Modeling meets Hyperspectral Remote Sensing [article]

Danfeng Hong and Wei He and Naoto Yokoya and Jing Yao and Lianru Gao and Liangpei Zhang and Jocelyn Chanussot and Xiao Xiang Zhu
2021 arXiv   pre-print
Compared to the convex models, non-convex modeling, which is capable of characterizing more complex real scenes and providing the model interpretability technically and theoretically, has been proven to  ...  Hyperspectral imaging, also known as image spectrometry, is a landmark technique in geoscience and remote sensing (RS).  ...  An Ever-Growing Relation between Non-convex Modeling and Interpretable AI in Hyperspectral Remote Sensing In recent years, a vast number of HS RS missions (e.g., MODIS, HypSEO, DESIS, Gaofen-5, EnMap,  ... 
arXiv:2103.01449v1 fatcat:jvo4pr5atvfb5kohpslvkhhmky

A Signal Processing Perspective on Hyperspectral Unmixing: Insights from Remote Sensing

Wing-Kin Ma, Jose M. Bioucas-Dias, Tsung-Han Chan, Nicolas Gillis, Paul Gader, Antonio J. Plaza, ArulMurugan Ambikapathi, Chong-Yung Chi
2014 IEEE Signal Processing Magazine  
B lind hyperspectral unmixing (HU), also known as unsupervised HU, is one of the most prominent research topics in signal processing (SP) for hyperspectral remote sensing [1], [2] .  ...  Research on this topic started in the 1990s in geoscience and remote sensing [3]-[7], enabled by technological advances in hyperspectral sensing at the time.  ...  The pure pixel concept actually came from the study of convex geometry (CG) of hyperspectral signals, where remote sensing researchers examined the special geometric structure of hyperspectral signals  ... 
doi:10.1109/msp.2013.2279731 fatcat:yltzlzhbmbfthfw4qim4bj2tjy

Non-convex Non-separable Sparse Nonnegative Matrix Factorization for Hyperspectral Unmixing

Fengchao Xiong, Jun Zhou, Jianfeng Lu, Yuntao Qian
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Over the past decade, nonnegative matrix factorization (NMF) has been utilized for this task, thanks to its good physical interpretation.  ...  Extensive experimental results on synthetic data and real-world data verify its utility when compared with several state-of-the-art approaches.  ...  The spectral bands significantly augment the discriminative ability, making HSI widely adopted in both remote sensing and computer vision tasks [1] , [2] .  ... 
doi:10.1109/jstars.2020.3028104 fatcat:cmw6t4kp6vbyzascay2yt33ot4

Robust Double Spatial Regularization Sparse Hyperspectral Unmixing

Fan Li, Shaoquan Zhang, Chengzhi Deng, Bingkun Liang, Jingjing Cao, Shengqian Wang
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
With the help of endmember spectral library, sparse unmixing techniques have been successfully applied to hyperspectral image interpretation.  ...  To this end, a pre-calculated spatial weighting factor is introduced to maintain the original spatial information of the hyperspectral image.  ...  INTRODUCTION Spectral unmixing is an important technique for hyperspectral image interpretation.  ... 
doi:10.1109/jstars.2021.3132164 fatcat:3z6y24m5tnd4rbiqax7etcbqle

Coupling NCA Dimensionality Reduction with Machine Learning in Multispectral Rock Classification Problems

Brian Bino Sinaice, Narihiro Owada, Mahdi Saadat, Hisatoshi Toriya, Fumiaki Inagaki, Zibisani Bagai, Youhei Kawamura
2021 Minerals  
Modestly put, hyperspectral imaging gathers electromagnetic signatures of the rocks in hundreds of spectral bands.  ...  In this study, we performed a 204-hyperspectral to 5-band multispectral reduction, because current production drones are limited to five multispectral bands sensors.  ...  However, researchers [11, 20, 26, 27] note that unlike the common PCA method, which is both convex and has an analytical solution, another key difference distinguishing the two is that NCA is a non-convex  ... 
doi:10.3390/min11080846 fatcat:eawf342xzzfc3kpgph2yel2uda

SpectralFormer: Rethinking Hyperspectral Image Classification with Transformers [article]

Danfeng Hong and Zhu Han and Jing Yao and Lianru Gao and Bing Zhang and Antonio Plaza and Jocelyn Chanussot
2021 arXiv   pre-print
Hyperspectral (HS) images are characterized by approximately contiguous spectral information, enabling the fine identification of materials by capturing subtle spectral discrepancies.  ...  Owing to their excellent locally contextual modeling ability, convolutional neural networks (CNNs) have been proven to be a powerful feature extractor in HS image classification.  ...  Zhu, “Interpretable hyperspectral artificial intelligence: When non- convex modeling meets hyperspectral remote  ... 
arXiv:2107.02988v2 fatcat:iw67o2iwhjafbhhrwogcswyk7u

Recent advances in techniques for hyperspectral image processing

Antonio Plaza, Jon Atli Benediktsson, Joseph W. Boardman, Jason Brazile, Lorenzo Bruzzone, Gustavo Camps-Valls, Jocelyn Chanussot, Mathieu Fauvel, Paolo Gamba, Anthony Gualtieri, Mattia Marconcini, James C. Tilton (+1 others)
2009 Remote Sensing of Environment  
Plaza et al. / Remote Sensing of Environment 113 (2009) S110-S122  ...  In this paper, we provide a seminal view on recent advances in techniques for hyperspectral image processing.  ...  This work has been supported by the European Community's Marie Curie Research Training Networks Programme under reference MRTN-CT-2006-035927, Hyperspectral Imaging Network (HYPER-I-NET).  ... 
doi:10.1016/j.rse.2007.07.028 fatcat:yhl3syflhjhvbhwsgiuvxht3lu

Distance metrics and band selection in hyperspectral processing with applications to material identification and spectral libraries

N. Keshava
2004 IEEE Transactions on Geoscience and Remote Sensing  
For example, ocean color remote sensing collects measurements in spectral regions where the relevant physical processes are observable.  ...  For example, sensors collecting data for ocean color remote sensing do not collect data beyond approximately 800nm because there is little, if any, reflected light at higher wavelengths.  ...  When using all bands, an unknown pixel requires K SAM angle comparisons in order to be classified. However, for both MDM and ADM, K-1 tests are performed that each require two SAM angle computations.  ... 
doi:10.1109/tgrs.2004.830549 fatcat:hgaggsiqurgtrhjdc7tgkzfx5y

Multiple Clustering Guided Nonnegative Matrix Factorization for Hyperspectral Unmixing

Wenhong Wang, Yuntao Qian, Hongfu Liu
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Spectral unmixing is an important technique for quantitatively analyzing hyperspectral remote sensing images.  ...  Specifically, in order to provide selfsupervised information to guide the NMF-based unmixing model, multiple clustering is integrated into the optimization process of NMF.  ...  Wang models are commonly used for HU: the linear spectral mixture model (LSMM) [1] and the non-linear spectral mixture model (NLSMM) [2] .  ... 
doi:10.1109/jstars.2020.3020541 fatcat:ccbmoy4vzvfoneriuhh43jux54

Development of soft computing and applications in agricultural and biological engineering

Yanbo Huang, Yubin Lan, Steven J. Thomson, Alex Fang, Wesley C. Hoffmann, Ronald E. Lacey
2010 Computers and Electronics in Agriculture  
Soft computing is a set of "inexact" computing techniques, which are able to model and analyze very complex problems.  ...  Yao and Tian (2003) proposed and tested a GA-based selective principal component analysis method using hyperspectral remote sensing data and ground reference data collected within a corn field for chlorophyll  ...  The results indicated that the ANN model with cross-learning using spectral information at 490, 570, 600, and 680 nm could be used to develop a practical remote sensing system to predict nitrogen content  ... 
doi:10.1016/j.compag.2010.01.001 fatcat:quszrg4vuzgcvkh2uglutnujdm

Knowledge Extracted from Copernicus Satellite Data

Dumitru Octavian, Schwarz Gottfried, Eltoft Torbjørn, Kræmer Thomas, Wagner Penelope, Hughes Nick, Arthus David, Fleming Andrew, Koubarakis Manolis, Datcu Mihai
2019 Zenodo  
The use of multi-or hyperspectral remote sensing data into soil monitoring and digital mapping can provide a large-scale survey, comprehensive and effective sites' monitoring and assess topsoil variables  ...  Remote sensing imagery is most widely used for water resources in remote area and monotonious regions. Several techniques were used to extract the glaciers lake surface using Landsat imagery.  ...  AI IN PHOTOGRAMMETRY AND REMOTE SENSING FOR DIGITAL EARTH Abstract This paper mainly describes the recent development of AI in photogrammetry and remote sensing at LIESMARS for Digital Earth.  ... 
doi:10.5281/zenodo.3941573 fatcat:zzifwgljifck5bpjnboetsftfu

Table of contents

2020 IEEE Transactions on Image Processing  
Yuan 5079 A Benchmark for Sparse Coding: When Group Sparsity Meets Rank Minimization ....................................... .................................................................... Z.  ...  Ni 5386 Multi-Granularity Canonical Appearance Pooling for Remote Sensing Scene Classification .............................. ...........................................................................  ... 
doi:10.1109/tip.2019.2940372 fatcat:h23ul2rqazbstcho46uv3lunku

Deep Learning for Geophysics: Current and Future Trends

Siwei Yu, Jianwei Ma
2021 Reviews of Geophysics  
Remote sensing imagery mainly includes optical images, hyperspectral images, and synthetic aperture radar (SAR) images.  ...  , since remote sensing is a common technique widely used in many areas.  ... 
doi:10.1029/2021rg000742 fatcat:6ifbu5izhzbxvhjfgmuh3vcznq

Current and near-term advances in Earth observation for ecological applications

Susan L. Ustin, Elizabeth M. Middleton
2021 Ecological Processes  
Lastly, we suggest instrument synergies that are likely to yield improved results when data are combined.  ...  Table 7 provides several open source packages and programs for analyzing remote sensing data today and more are being added all the time.  ...  This Archive category provides current users of remote sensing data with past instruments because of many requirements for time series analyses.  ... 
doi:10.1186/s13717-020-00255-4 pmid:33425642 pmcid:PMC7779249 fatcat:tojcb7xw3reajgwgci46zsuqk4

Table of contents

2021 ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
.............................. 4235 FOR DEPRESSION DETECTION Meng Niu, Kai Chen, Qingcai Chen, Lufeng Yang, Harbin Institute of Technology, Shenzhen, China MMSP-5.2: WHEN FACE RECOGNITION MEETS OCCLUSION  ...  IMAGING IVMSP-14.1: NMF-SAE: AN INTERPRETABLE SPARSE AUTOENCODER FOR ............................................. 1785 HYPERSPECTRAL UNMIXING Fengchao Xiong, Nanjing University of Science and Technology  ... 
doi:10.1109/icassp39728.2021.9414617 fatcat:m5ugnnuk7nacbd6jr6gv2lsfby
« Previous Showing results 1 — 15 out of 51 results