themes of data fusion and feature selection/extraction, cover almost 3/4 of ... However, the rising themes of .... Devis Tuia S'07âM'09âSM'15) was born in Men-.
IEEE JSTARS, VOL. XX, NO. Y, MONTH Z 2015. PREPRINT. FOR THE FULL VERSION, PLEASE REFER TO IEEE XPLORE.
1
Foreword to the Special Issue on Hyperspectral Image and Signal Processing This special issue presents advances in signal and image processing related to hyperspectral remote sensing (or imaging spectroscopy). We present the seventy-eight papers selected, which distill the state of the art and the advances in hyperspectral image analysis and its applications. This special issue of JSTARS follows the 6th workshop on hyperspectral image and signal processing (WHISPERS) held in Lausanne (Switzerland) in June 2014. It contains both extended versions of contributions presented in WHISPERS and regular papers submitted to JSTARS dealing with contemporary aspects of hypersepctral image and signal processing. Main topics show continuity. At a first glimpse, the papers contained in this special issue maintain the methodological flavour of both WHISPERS and the previous special issues [1]–[3]. The main topics remain classification, unmixing, spectral-spatial methods and sparse methods, as one can see in the word cloud extracted form the titles of the papers published in this issue (Fig. 1). This is confirmed by the rough classification of the main topics1 of the papers, which is reported in the top panel of Fig. 2: classification and unmixing are still the most present ones and, together with the connecting themes of data fusion and feature selection/extraction, cover almost 3/4 of the keywords. However, the rising themes of high performance computing, regression and sparse processing seem to make their way into the favourite topics represented and will surely increase their presence in the future. Hence, the hypersepctral image/signal processing community seems to stick to its favourite themes, while enlarging its vision thanks to the advances in computing power [4] and availability of data [5]. New applications are being tackled, but is it fast enough? From the application summary in the bottom panel of Fig. 2, one can observe that roughly 50% of the papers in the issue validate the proposed methodology on benchmark datasets (e.g. AVIRIS Indian Pines, Salinas, Cuprite or KSC, Hyperion Botswana): despite the need to test the proposed algorithms against well-known benchmarks, this is somehow alarming, since these benchmark datasets are nowadays solved with accuracy of around 100% (in the case of classification) or lack a quantitative benchmarking (in the case of Cuprite). We see here the danger of falling in the trap of trying to satisfy the validatiaon data supplied as good as possible rather than looking for generalization: following this path may create a self-sustaining community of methods developers losing touch with the applications and data providers. Even if we recognize the long lasting legacy and importance of these datasets, we 1 Each
content.
paper was assigned to a number of subtopics that matched with its
think there is a great and urgent need for new benchmark datasets, where methodological advances can really be assessed, coming with standardized measures of performance. These datasets must be available openly and reflect the variety of situations and complexity that is found in real data: now that hyperspectral sensors have hit the market of consumers (many systems are now available for aerial or drone-based imaging spectroscopy) and that there is interest from the usersside, the image / signal processing community must be able to provide showcase exemples of the potential to new generation HSI sensors. But to do so, it needs to move to datasets with realistic challenges such as large datasets (billions of samples, mosaicks), absolute/relative spectral consistency (via physical approaches or domain adaptation) or multitemporal / multiresolution monitoring. But there are already signals in the good direction: around 12% of the papers already use new benchmark data provided, for instance, through the Image Analysis and Data Fusion Technical Committee of the IEEE-GRSS2 [6], [7]. These new benchmarks start to address these problems and we hope they are only a first step in the direction of new challenging hyperspectral data where image and signal processing methods can be tested in a fair and objective way. Finally, we also observe several papers dealing with challenging applications, including the study of functional vegetation traits and the application of hyperspectral imagery in forestry. We hope that the purely methodological community will find interest in these applications and consider them as a set of new opportunities for their new developments. We also hope that this will happen jointly with the application-oriented 2 http://www.grss-ieee.org/community/technical-committees/data-fusion/
Fig. 1. WordCloud of the words appearing in the titles of the paper in the special issue. Size and color relate to the frequency of appearance of the word (created with Worditout.com).
2
IEEE JSTARS, VOL. XX, NO. Y, MONTH Z 2015. PREPRINT. FOR THE FULL VERSION, PLEASE REFER TO IEEE XPLORE.
topics is as wide as the different methodological propositions: we believe that this issue constitutes a contemporary view on the field and on its actors. We thank all the authors that contributed, as well as the numerous reviewers, who offered their time and effort to make this issue happen. The guest editors, aaaaaa D. Tuia, S. Lopez, M. Schaepman, J. Chanussot. R EFERENCES
Fig. 2. Themes (top) and applications (bottom) in the special issue.
community and not separately or in parallel. We are reinforcing the links with machine learning / computer vision. As a last discussion item, we also note that the hyperspectral image / signal processing community is keeping the pace with the computer vision / machine learning ones, as very recent topics such as deep neural networks or structured learning are also represented in the issue. We think that staying close to these communities is important, but we also raise the concern of not just remaining spectators of what happens in machine learning / vision, but also try to act and be present there, to show there that hyperspectral image processing is an interesting problem for vision and machine learning. Recent papers in top conferences in vision / machine learning show that such an active role is possible (only at the last CVPR, 5 papers dealt with hyperspectral imaging [8]– [12]) and we encourage the remote sensing hyperspectral community to be proactive and not only be consumers of methods coming from these fields. We believe that the specific characteristics of hyperspectral data and the considered applications can reach out a larger audience beyond remote sensing. Said that, we hope you will enjoy reading this special issue and find new inspiring and refreshing ideas. The range of
[1] Q.D. Du, L.Z. Zhang, B.Z. Zhang, X.T. Tong, P.D. Du, and J.C. Chanussot, “Foreword to the special issue on hyperspectral remote sensing: Theory, methods, and applications,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 6, no. 2, pp. 459–465, April 2013. [2] A. Zare, J. Bolton, J. Chanussot, and P. Gader, “Foreword to the special issue on hyperspectral image and signal processing,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 7, no. 6, pp. 1841–1843, June 2014. [3] A. Plaza, J. M. Bioucas-Dias, A. Simic, and W. J. Blackwell, “Foreword to the special issue on hyperspectral image and signal processing,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 5, no. 2, pp. 347–353, April 2012. [4] S. Lopez, T. Vladimirova, C. Gonz´alez, J. Resano, D. Mozos, and A. Plaza, “The promise of reconfigurable computing for hyperspectral imaging onboard systems: A review and trends,” Proc. of IEEE, vol. 101, no. 3, pp. 698–722, 2013. [5] G. Camps-Valls, D. Tuia, L. Bruzzone, and J. A. Benediktsson, “Advances in hyperspectral image classification,” IEEE Signal Proc. Mag., vol. 31, pp. 45–54, 2014. [6] C. Debes, A. Merentitis, R. Heremans, J. Hahn, N. Frangiadakis, T. van Kasteren, Wenzhi Liao, R. Bellens, A. Pizurica, S. Gautama, W. Philips, S. Prasad, Qian Du, and F. Pacifici, “Hyperspectral and lidar data fusion: Outcome of the 2013 grss data fusion contest,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 7, no. 6, pp. 2405–2418, June 2014. [7] W. Liao, X. Huang, F. Van Collie, A. Gautama, W. Philips, H. Liu, T. Zhu, M. Shimoni, G. Moser, and D. Tuia, “Processing of thermal hyperspectral and digital color cameras: outcome of the 2014 data fusion contest,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, in press. [8] Y. Zheng, I. Sato, and Y. Sato, “Illumination and reflectance spectra separation of a hyperspectral image meets low-rank matrix factorization,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015. [9] L. Zhang, W. Wei, Y. Zhang, C. Tian, and F. Li, “Reweighted laplace prior based hyperspectral compressive sensing for unknown sparsity,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015. [10] N. Akhtar, F. Shafait, and A. Mian, “Bayesian sparse representation for hyperspectral image super resolution,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015. [11] L. Wang, Z. Xiong, D. Gao, G. Shi, W. Zeng, and F. Wu, “High-speed hyperspectral video acquisition with a dual-camera architecture,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015. [12] S. Feng, M. F. Duarte, and M. Parente, “Universality of wavelet-based non-homogeneous hidden markov chain model features for hyperspectral signatures,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June 2015.
Foreword to the Special Issue on Hyperspectral Image and Signal Processing
Devis Tuia S’07–M’09–SM’15) was born in Mendrisio, Switzerland, in 1980. He received a diploma in Geography at the University of Lausanne (UNIL) in 2004, the Master of Advanced Studies in Environmental Engineering at the Federal Institute of Technology of Lausanne (EPFL) in 2005 and a Ph.D. in Environmental Sciences at UNIL in 2009. He was then a visiting postdoc researcher at the University of Val´encia, Val´encia, Spain and the University of Colorado, Boulder, CO, USA. He then worked as Senior Research Associate at EPFL under a Swiss National Foundation program. Since 2014, he is Assistant Professor at the Department of Geography of the University of Zurich. His research interests include the development of algorithms for information extraction and data fusion of remote sensing images using machine learning algorithms. Dr. Tuia serves as a Cochair of the I MAGE A NALYSIS AND DATA F USION T ECHNICAL C OMMITTEE of the IEEE GRSS. He is an Associate Editor of the IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATION AND REMOTE SENSING . Visit http://devis.tuia.googlepages.com/ for more information.
Sebastian Lopez (M‘08-SM‘15) was born in Las Palmas de Gran Canaria, Spain, in 1978. He received the M.S. degree in Electronic Engineering by the University of La Laguna in 2001, obtaining regional and national awards for his CV during his degree. He got his PhD degree in Electronic Engineering by the University of Las Palmas de Gran Canaria in 2006, where he is currently an Associate Professor, developing his research activities at the Integrated Systems Design Division of the Institute for Applied Microelectronics (IUMA). He is currently an Associate Editor of the IEEE J OURNAL OF S ELECTED T OPICS IN A PPLIED E ARTH O BSERVATIONS AND R EMOTE S ENSING (JSTARS) an AdCom member of the Spanish Chapter of the IEEE GRSS. He also was an associate editor of the IEEE T RANSACTIONS ON C ONSUMER E LEC TRONICS from 2008 to 2013. Additionally, he currently serves as an active reviewer of the IEEE JSTARS, IEEE T RANSACTIONS ON G EOSCIENCE AND R EMOTE S ENSING , IEEE G EOSCIENCE AND R EMOTE S ENSING L ETTERS, IEEE T RANSACTIONS ON C IRCUITS AND S YSTEMS FOR V IDEO T ECHNOLOGY, the Journal of Real Time Image Processing, Microprocessors and Microsystems: Embedded Hardware Design (MICPRO), and the IET Electronics Letters, among others. He is also a program committee member of different international conferences including the SPIE Conference on Satellite Data Compression, Communication and Processing, IEEE Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), and SPIE Conference of High Performance Computing in Remote Sensing. Furthermore, he acted as one of the program chairs at the last two aforementioned conferences for their 2014 editions and will be the program chair of the SPIE Conference of High Performance Computing in Remote Sensing for its 2015 edition. Moreover, he was the guest editor of the special issue entitled “Design and Verification of Complex Digital Systems” that was published in 2011 at the Elsevier Microprocessors and Microsystems: Embedded Hardware Design (MICPRO) journal. He has published more than 80 papers in international journals and conferences. His current research interests include real-time hyperspectral imaging, reconfigurable architectures, high-performance computing systems, and image and video processing.
3
Michael Schaepman (M‘05 SM‘07) received the M.Sc. and Ph.D. degrees in geography from the University of Z¨urich (UZH), Z¨urich, Switzerland, in 1993 and 1998, respectively. In 1999, he spent his Postdoctoral time with the Optical Sciences Center, University of Arizona, Tucson, AZ, USA. In 2000, he was appointed as a Project Manager of the European Space Agency APEX Spectrometer. In 2003, he accepted the position of a Full Chair of Geoinformation Science and Remote Sensing with Wageningen University, Wageningen, The Netherlands. In 2009, he was appointed as the Full Chair of Remote Sensing with UZH, where he is currently heading the Remote Sensing Laboratories, Department of Geography. His research interests include computational Earth sciences using remote sensing and physical models, with particular focus on the land?atmosphere interface using IS.
Jocelyn Chanussot (M‘04-SM‘04-F‘12) received the M.Sc. degree in electrical engineering from the Grenoble Institute of Technology (Grenoble INP), Grenoble, France, in 1995, and the Ph.D. degree from Savoie University, Annecy, France, in 1998. In 1999, he was with the Geography Imagery Perception Laboratory for the Delegation Generale de l’Armement (DGA-French National Defense Department). Since 1999, he has been with Grenoble INP, Saint-Martin-d’H`eres, France, where he was an Assistant Professor from 1999 to 2005, an Associate Professor from 2005 to 2007, and is currently a Professor of Signal and Image Processing. He is a Member of the Institut Universitaire de France (20122017). Since 2013, he has been an Adjunct Professor with the University of Iceland, Reykjavk, Iceland. His research interests include image analysis, multicomponent image processing, nonlinear filtering, and data fusion in remote sensing. Dr. Chanussot is the founding President of IEEE Geoscience and Remote Sensing French chapter (2007-2010), which received the 2010 IEEE GRS-S Chapter Excellence Award. He was the Corecipient of the NORSIG 2006 Best Student Paper Award, the IEEE GRSS 2011 Symposium Best Paper Award, the IEEE GRSS 2012 Transactions Prize Paper Award, and the IEEE GRSS 2013 Highest Impact Paper Award. He was a Member of the IEEE Geoscience and Remote Sensing Society AdCom (2009-2010), in charge of membership development. He was the General Chair of the first IEEE GRSS Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing (WHISPERS). He was the Chair (2009-2011) and Cochair of the GRS Data Fusion Technical Committee (2005-2008). He was a Member of the Machine Learning for Signal Processing Technical Committee of the IEEE Signal Processing Society (2006?2008) and the Program Chair of the IEEE International Workshop on Machine Learning for Signal Processing, (2009). He was an Associate Editor for the IEEE Geoscience and Remote Sensing Letters (2005?2007) and for Pattern Recognition (2006-2008). Since 2007, he has been an Associate Editor for the IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING. Since 2011, he has been the Editor-in-Chief of the IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING. In 2013, he was a Guest Editor for the PROCEEDINGS OF THE IEEE and in 2014, a Guest Editor for the IEEE S IGNAL P ROCESSING M AGAZINE. He is a Member of the Institut Universitaire de France (2012-2017).