Skip to Main content Skip to Navigation
Conference papers

Combining multiscale features for classification of hyperspectral images: a sequence based kernel approach

Yanwei Cui 1 Laëtitia Chapel 1 Sébastien Lefèvre 1
1 OBELIX - Environment observation with complex imagery
IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE, UBS - Université de Bretagne Sud
Abstract : Nowadays, hyperspectral image classification widely copes with spatial information to improve accuracy. One of the most popular way to integrate such information is to extract hierarchical features from a multiscale segmentation. In the classification context, the extracted features are commonly con-catenated into a long vector (also called stacked vector), on which is applied a conventional vector-based machine learning technique (e.g. SVM with Gaussian kernel). In this paper , we rather propose to use a sequence structured kernel: the spectrum kernel. We show that the conventional stacked vector-based kernel is actually a special case of this kernel. Experiments conducted on various publicly available hyper-spectral datasets illustrate the improvement of the proposed kernel w.r.t. conventional ones using the same hierarchical spatial features.
Document type :
Conference papers
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01320012
Contributor : Sébastien Lefèvre <>
Submitted on : Wednesday, November 13, 2019 - 5:51:26 PM
Last modification on : Friday, July 10, 2020 - 4:19:40 PM

File

whispers2016.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01320012, version 1

Citation

Yanwei Cui, Laëtitia Chapel, Sébastien Lefèvre. Combining multiscale features for classification of hyperspectral images: a sequence based kernel approach. Workshop on Hyperspectral Image and Signal Processing : Evolution in Remote Sensing (WHISPERS), 2016, Los Angeles, United States. ⟨hal-01320012⟩

Share

Metrics

Record views

932

Files downloads

136