Scientists at the University of Iceland recently received an international award for the best science article in the International Journal of Image and Data Fusion for 2012-2013.
The paper was published in Volume 3, issue 3 in 2012 and is entitled Classification of hyperspectral data using extended attribute profiles based on supervised and unsupervised feature extraction techniques.
The Best Paper Award for International Journal of Image and Data Fusion is presented once every two years to recognise outstanding papers published in the journal. Its purpose is to encourage researchers and experts in this field to share original research through theInternational Journal of Image and Data Fusion. The winner of the award will be selected at the sole discretion of the Editor-in-Chief and his Editorial Board, who make up the Award Committee. The candidates for the award will adhere to the requirements of good science, ethics and best practice.
The paper is based on research at the University of Iceland, managed by Jón Atli Benediktsson, Pro Rector of Academic Affairs and Professor at the Faculty of Electrical and Computer Engineering. Benediktsson worked on the research with former employees of the University, Prashanth Reddy Marpu, who was working as a Post doc, and Mattia Pedergnana, a specialist; both at the Faculty of Electrical and Computer Engineering. Mauro Dalla Maura who completed a joint doctoral degree from the University of Iceland and the University of Trento in 2011 also worked on the research as did Stijn Peeters, a masters student at the University of Antwerpen, Belgium, but he took part of his graduate studies at the University of Iceland under the supervision of Jón Atli Benediktsson.
The paper was written in collaboration with Lorenzo Bruzzone, Department of Information Engineering and Computer Science, University of Trento, Trento, Italy. The paper discusses the analysis of remote sensing images using mathematical morphology for image retrieval from remote sensing images. With the development of efficient algorithms to construct the profiles for large datasets, such methods are becoming even more relevant. When dealing with hyperspectral imagery, the profiles are traditionally built on the first few principal components computed from the data, as is stated in the paper’s abstract.
The corresponding author, on behalf of his co-authors, has received a certificate endorsed by both the Editor-in-Chief of the journal and Taylor & Francis, along with the Best Paper Prize award of GBP £500.
The website of International Journal of Image and Data Fusion.