On July 6, 2018, severe damage and causalities occurred due to the torrential rain in western Japan. The Geoinformatics Unit from RIKEN-AIP and the International Research Institute of Disaster Science (IRIDeS) conducted a preliminary damage mapping using satellite remote sensing imagery (ALOS-2) acquired from the affected areas. The ALOS-2 data used in this analysis was provided by the large-scale disaster satellite image analysis support team of the National Aerospace Exploration Agency of Japan.

We are looking for motivated and skilled internship students who want to work on machine learning in remote sensing. Thesis projects are also welcomed.

Our work will be presented in seven contributions (four posters and three oral presentations) at <a href="https://igarss2018.org/default.asp">IGARSS 2018</a> in Valencia, Spain!! Bruno and Naoto will attend the conference. More details of our contributions are listed below.<br><br> Title: DAMAGE MAPPING AFTER THE 2017 PUEBLA EARTHQUAKE IN MEXICO USING HIGH-RESOLUTION ALOS2 PALSAR2 DATA<br> Topic: Land Applications: Urban and Built Environment<br> Session: MOP2.PL: Urban and Built Environment I<br> Time: Monday, July 23, 15:50 - 16:50<br> Authors: Bruno Adriano, Shunichi Koshimura, Sadra Karimzadeh, Masashi Matsuoka, Magaly Koch<br><br> Title: SMALL SIZE CLASS PRESERVING CLASSIFICATION BASED ON SEGMENTATION FOR HYPERSPECTRAL DATA<br> Topic: Data Analysis Methods (Optical, Multispectral, Hyperspectral, SAR): Classification and Clustering<br> Session: TUP1.PH: Classification of Hyperspectral Data<br> Time: Tuesday, July 24, 10:10 - 11:10<br> Authors: Tatsuya Yamada, Junshi Xia, Akira Iwasaki<br><br> Title: SUPERPIXEL BASED DIMENSION REDUCTION FOR HYPERSPECTRAL IMAGERY<br> Topic: Data Analysis Methods (Optical, Multispectral, Hyperspectral, SAR): Classification and Clustering<br> Session: TUP1.PG: Spectral-Spatial Approaches in Hyperspectral Remote Sensing<br> Time: Tuesday, July 24, 10:10 - 11:10<br> Authors: Huilin Xu, Hongyan Zhang, Wei He, Liangpei Zhang<br><br> Title: MULTIPLE SOURCES DATA FUSION VIA DEEP FOREST<br> Topic: Invited Sessions: Data Fusion<br> Session: TU3.R7: Data Fusion I<br> Time: Tuesday, July 24, 14:10 - 15:50<br> Authors: Junshi Xia, Zuheng Ming, Akira Iwasaki<br><br> Title: THE DATA FUSION CONTEST 2018: ADVANCED MULTI-SENSOR OPTICAL REMOTE SENSING FOR URBAN LAND USE AND LAND COVER CLASSIFICATION<br> Topic: Invited Sessions: IEEE GRSS Data Fusion Contest<br> Session: WE1.R5: IEEE GRSS Data Fusion Contest<br> Time: Wednesday, July 25, 08:30 - 10:10<br> Authors: Naoto Yokoya, Bertrand Le Saux, Ronny Hänsch, Saurabh Prasad<br><br> Title: A COMPARATIVE STUDY OF FUSION-BASED CHANGE DETECTION METHODS FOR MULTI-BAND IMAGES WITH DIFFERENT SPECTRAL AND SPATIAL RESOLUTIONS<br> Topic: Data Analysis Methods (Optical, Multispectral, Hyperspectral, SAR): Change Detection and Multi-Temporal Analysis<br> Session: WEP1.PJ: Techniques for Multi-temporal Optical Image Analysis<br> Time: Wednesday, July 25, 10:10 - 11:10<br> Authors: Vinicius Ferraris, Naoto Yokoya, Nicolas Dobigeon, Marie Chabert<br><br> Title: BOOSTING FOR DOMAIN ADAPTATION EXTREME LEARNING MACHINES FOR HYPERSPECTRAL IMAGE CLASSIFICATION<br> Topic: Data Analysis Methods (Optical, Multispectral, Hyperspectral, SAR): Classification and Clustering<br> Session: WE4.R1: Learning and Domain Adaptation<br> Time: Wednesday, July 25, 16:50 - 18:30<br> Authors: Junshi Xia, Naoto Yokoya, Akira Iwasaki<br><br><br> Furthermore, we will organize the following invited sessions:<br><br> <a href="https://igarss2018.org/Papers/PublicSessionIndex3.asp?Sessionid=1371">TU3.R7: Data Fusion I</a><br> <a href="https://igarss2018.org/Papers/PublicSessionIndex3.asp?Sessionid=1372">TU4.R7: Data Fusion II</a><br> <a href="https://igarss2018.org/Papers/PublicSessionIndex3.asp?Sessionid=1328">WE1.R5: IEEE GRSS Data Fusion Contest</a><br><br>

Pleased to announce that our paper entitled "Joint & Progressive Learning from High-Dimensional Data for Multi-Label Classification," written by Danfeng Hong, Naoto Yokoya, Jian Xu, and Xiaoxiang Zhu was accepted to <a href="https://eccv2018.org/">European Conference on Computer Vision 2018</a>!! This paper proposes a linearized subspace learning technique to improve explainability, generalization, and cost-effectiveness. Experiments on hyperspectral remote sensing datasets and face datasets demonstrate the superiority and effectiveness of the proposed method in comparison with previous state-of-the-art methods.

Dr. Tien Dat Pham gave a special lecture entitled “Monitoring Mangrove forest changes Using GIS, Remote Sensing Data and Machine Learning Techniques for Implementation of REDD+ Policies in Vietnam” on Remote Sensing/GIS course at United Nations University (UNU).

Our paper entitled "Open data for global multimodal land use classification: Outcome of the 2017 IEEE GRSS Data Fusion Contest" has been published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.<br><br> The paper summarizes the outcome of the 2017 IEEE GRSS Data Fusion Contest, which was focused on local climate zones classification using multimodal data (i.e., multitemporal Landsat-8, Sentinel-2, and OpenStreetMap).<br><br> More details are reported in the following open access paper:<br> N. Yokoya, P. Ghamisi, J. Xia, S. Sukhanov, R. Heremans, I. Tankoyeu, B. Bechtel, B. Le Saux, G. Moser, and D. Tuia, "<a href="https://ieeexplore.ieee.org/document/8338367/">Open data for global multimodal land use classification: Outcome of the 2017 IEEE GRSS Data Fusion Contest</a>,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 11, no. 5, pp. 1363-1377, 2018.