Naslov
Tensor-based Offset-Sparsity Decomposition for Hyperspectral Image Classification

Autori
Tian, Long ; Du, Qian ; Younan Nicolas ; Kopriva, Ivica

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Geoscience and Remote Sensing Symposium (IGARSS), 2017 IEEE International / Tjuatja, Saibun ; Kunkee, David - Piscataway (NJ) : Institute of Electrical and Electronics Engineers (IEEE), 2017, 3656-3659

ISBN
978-1-5090-4951-6

Skup
Geoscience and Remote Sensing Symposium (IGARSS), 2017 IEEE International

Mjesto i datum
Fort Worth (TX), Sjedinjene Američke Države, 23.07.2017. - 28.07.2017

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Tensor ; low-rank recovery ; spatial-spectral segmentation ; classification ; hyperspectral imagery

Sažetak
In this paper, the tensor-based offset-sparsity decomposition (TOSD) method, or low-rank and sparse decomposition, is applied to hyperspectral imagery, where the low-rank tensor is considered to be enhanced or pruned data and used for classification. In the tensor form of dataset, all the information of the original 3D data cube, includes spatial and spectral information, can be better reserved. To make the low-rank assumption more possibly true, spatial and spectral segmentations are conducted in a preprocessing step for the TOSD. The experimental results demonstrate the TOSD offers better performance than the matrix-based one, and the spatial-spectral segmentation can further improve the performance.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Računarstvo

Napomena
Rad je rezultat zanstvene suradnje Ivice Koprive sa Instituta Ruđer Bošković i profesorice Qian Du sa Mississippi State University, SAD. Rad nije dio formalno prihvaćenog znanstvenog projekta.

POVEZANOST RADA