Best Paper for the period 2012-2015
The U.V. Helava Award, sponsored by Elsevier B.V. and Leica Geosystems AG, is a prestigious ISPRS Award, which was established in 1996 to encourage and stimulate submission of high quality scientific papers by individual authors or groups to the ISPRS Journal, to promote and advertise the Journal, and to honour the outstanding contributions of Dr. Uuno V. Helava to research and development in photogrammetry and remote sensing.
The Award is presented to authors of the best paper, written in English and published exclusively in the ISPRS Journal during the four-year period from January of a Congress year, to December of the year prior to the next Congress.
The award consists of a monetary grant of 10,000 SFr., certificates and a silver plaque. It is sponsored by Elsevier B.V. and Hexagon Geosystems, while the Institute of Photogrammetry and Remote Sensing (Prof. Henrik Haggrén), Helsinki University of Technology (the University where Helava studied) paid half the costs for the silver plaque. The plaque was designed with care and enthusiasm by the 1980-88 ISPRS Technical Commission III President, Einari Kilpelä, previously professor at the Helsinki University of Technology.
A five-member Jury, comprising experts of high scientific standing, whose expertise covers the main topics included in the scope of the Journal, evaluates the papers. For each year of the four-year evaluation period, the best paper is selected, and among these four papers, the one to receive the U.V. Helava Award. The award winning paper is:
"Multiclass feature learning for hyperspectral image classification: Sparse and hierarchical solutions"
by Devis Tuia1, Rémi Flamary2 and Nicolas Courty3
1 University of Zurich, Switzerland
2 Université de Nice Sofia Antipolis, France
3 Université de Bretagne Sud/IRISA, France
published in volume 105, July 2015, pp. 272-285, http://dx.doi.org/10.1016/j.isprsjprs.2015.01.006
Jury's rationale for the paper selection
The winning paper addresses the problem of high dimensionality in hyperspectral image classification. Presenting well the shortcomings of existing approaches, the authors solve the problem of filter bank selection by choosing only those that contribute to improving land cover classification. Their work features a number of innovations including the use of multi-class logistic regression, group-lasso regularization that allows information sharing between thematic classes and automatic filter selection. Their new approach was tested thoroughly on both agricultural and urban scenes. The Jury was impressed by the authors’ innovative methodology, finding the performance-based feature selection technique to be very novel yet still tractable. They felt this well-written contribution represents a genuine scientific advance in hyperspectral image classification and, therefore, very deserving of the U.V. Helava Award for 2012-2015.
The fifth U.V. Helava Award will be presented at the 23rd ISPRS Congress in Prague, 12-19 July 2016 by Chen Jun, ISPRS President, and representatives of the sponsors.
On behalf of the ISPRS and the U.V. Helava Award Jury, I would like to congratulate the authors for this distinction and thank them for their contribution. I would also like to thank the sponsors of the Award, and the Jury members for their thorough evaluations.
ISPRS Journal of Photogrammetry and Remote Sensing,
The University of Calgary, Canada
E-mail address: email@example.com