Application of unsupervised nearest-neighbor density-based approaches to sequential dimensionality reduction and clustering of hyperspectral images

Claude Cariou; Kacem Chehdi

doi:10.1117/12.2325530

Translator Disclaimer

9 October 2018 Application of unsupervised nearest-neighbor density-based approaches to sequential dimensionality reduction and clustering of hyperspectral images

Claude Cariou; Kacem Chehdi

Author Affiliations +

Proceedings Volume 10789, Image and Signal Processing for Remote Sensing XXIV; 107890H (2018) https://doi.org/10.1117/12.2325530
Event: SPIE Remote Sensing, 2018, Berlin, Germany

Abstract

In this communication, we address the problem of unsupervised dimensionality reduction (DR) for hyperspectral images (HSIs), using nearest-neighbor density-based (NN-DB) approaches. Dimensionality reduction is an important tool in the HSI processing chain, aimed at reducing the high redundancy among the HSI spectral bands, while preserving the maximum amount of relevant information for further processing. Basically, the idea is to formalize DR as the process of partitioning the spectral bands into coherent band sets. Two DR schemes can be set up directly, one based on band selection, and the other one based on band averaging. Another scheme is proposed here, based on compact band averaging. Experiments are conducted with hyperspectral images composed of an AISA Eagle HSI issued from our acquisition platform, and the AVIRIS Salinas HSI. We evaluate the efficiency of the reduced HSIs for final classification results under the three schemes, and compare them to the classification results without reduction. We show that despite a high dimensionality reduction (< 8% of the bands left), the clustering results provided by NN-DB methods remain comparable to the ones obtained without DR, especially for GWENN in the band averaging case. We also compare the classification results obtained after applying other unsupervised or semi-supervised DR schemes, based either on band selection or band averaging, and show the superiority of the proposed DR scheme.

Citation Download Citation

Claude Cariou and Kacem Chehdi "Application of unsupervised nearest-neighbor density-based approaches to sequential dimensionality reduction and clustering of hyperspectral images", Proc. SPIE 10789, Image and Signal Processing for Remote Sensing XXIV, 107890H (9 October 2018); https://doi.org/10.1117/12.2325530

PROCEEDINGS
12 PAGES

DOWNLOAD PDF SAVE TO MY LIBRARY

SHARE

GET CITATION

< Previous Article

|

Next Article >

KEYWORDS

Hyperspectral imaging

Principal component analysis

Data processing

Data storage

Denoising

Electronics

Fuzzy logic

Show All Keywords

RELATED CONTENT

Fast techniques for nonlinear mapping of hyperspectral data
Proceedings of SPIE (March 17 2017)

Investigation of spectral screening techniques for independent component analysis based...
Proceedings of SPIE (September 23 2003)

Hyperspectral image feature extraction accelerated by GPU
Proceedings of SPIE (October 24 2012)

Research and design of high speed mass image storage system
Proceedings of SPIE (August 04 2009)

Fast hyperspectral data processing methods
Proceedings of SPIE (June 01 2001)

Band prioritization for hyperspectral imagery
Proceedings of SPIE (September 01 2006)

Development of a forestry government agency enterprise GIS system ...
Proceedings of SPIE (November 05 2008)

Subscribe to Digital Library

Receive Erratum Email Alert

Show All Keywords

Keywords/Phrases

Search In:

Publication Years