Assessment of Component Selection Strategies in Hyperspectral Imagery
Hyperspectral imagery (HSI) integrates many continuous and narrow bands that cover different regions of the electromagnetic spectrum. However, the main challenge is the high dimensionality of HSI data due to the ’Hughes’ phenomenon. Thus, dimensionality reduction is necessary before applying classif...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-12-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/19/12/666 |
| _version_ | 1798026982145916928 |
|---|---|
| author | Edurne Ibarrola-Ulzurrun Javier Marcello Consuelo Gonzalo-Martin |
| author_facet | Edurne Ibarrola-Ulzurrun Javier Marcello Consuelo Gonzalo-Martin |
| author_sort | Edurne Ibarrola-Ulzurrun |
| collection | DOAJ |
| description | Hyperspectral imagery (HSI) integrates many continuous and narrow bands that cover different regions of the electromagnetic spectrum. However, the main challenge is the high dimensionality of HSI data due to the ’Hughes’ phenomenon. Thus, dimensionality reduction is necessary before applying classification algorithms to obtain accurate thematic maps. We focus the study on the following feature-extraction algorithms: Principal Component Analysis (PCA), Minimum Noise Fraction (MNF), and Independent Component Analysis (ICA). After a literature survey, we have observed a lack of a comparative study on these techniques as well as accurate strategies to determine the number of components. Hence, the first objective was to compare traditional dimensionality reduction techniques (PCA, MNF, and ICA) in HSI of the Compact Airborne Spectrographic Imager (CASI) sensor and to evaluate different strategies for selecting the most suitable number of components in the transformed space. The second objective was to determine a new dimensionality reduction approach by dividing the CASI HSI regarding the spectral regions covering the electromagnetic spectrum. The components selected from the transformed space of the different spectral regions were stacked. This stacked transformed space was evaluated to see if the proposed approach improves the final classification. |
| first_indexed | 2024-04-11T18:44:16Z |
| format | Article |
| id | doaj.art-77c7025376ca4165814daf3dad2076b2 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-04-11T18:44:16Z |
| publishDate | 2017-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-77c7025376ca4165814daf3dad2076b22022-12-22T04:08:52ZengMDPI AGEntropy1099-43002017-12-01191266610.3390/e19120666e19120666Assessment of Component Selection Strategies in Hyperspectral ImageryEdurne Ibarrola-Ulzurrun0Javier Marcello1Consuelo Gonzalo-Martin2Instituto de Oceanografía y Cambio Global, IOCAG, Universidad de Las Palmas de Gran Canaria, ULPGC, Parque Científico Tecnológico Marino de Taliarte, s/n, 35214 Telde, SpainInstituto de Oceanografía y Cambio Global, IOCAG, Universidad de Las Palmas de Gran Canaria, ULPGC, Parque Científico Tecnológico Marino de Taliarte, s/n, 35214 Telde, SpainCenter of Biomedical Technology, Universidad Politécnica de Madrid, UPM, Campus de Montegancedo, Pozuelo de Alarcón, 28223 Madrid, SpainHyperspectral imagery (HSI) integrates many continuous and narrow bands that cover different regions of the electromagnetic spectrum. However, the main challenge is the high dimensionality of HSI data due to the ’Hughes’ phenomenon. Thus, dimensionality reduction is necessary before applying classification algorithms to obtain accurate thematic maps. We focus the study on the following feature-extraction algorithms: Principal Component Analysis (PCA), Minimum Noise Fraction (MNF), and Independent Component Analysis (ICA). After a literature survey, we have observed a lack of a comparative study on these techniques as well as accurate strategies to determine the number of components. Hence, the first objective was to compare traditional dimensionality reduction techniques (PCA, MNF, and ICA) in HSI of the Compact Airborne Spectrographic Imager (CASI) sensor and to evaluate different strategies for selecting the most suitable number of components in the transformed space. The second objective was to determine a new dimensionality reduction approach by dividing the CASI HSI regarding the spectral regions covering the electromagnetic spectrum. The components selected from the transformed space of the different spectral regions were stacked. This stacked transformed space was evaluated to see if the proposed approach improves the final classification.https://www.mdpi.com/1099-4300/19/12/666remote sensinghyperspectral sensorfeature-extractiontexture measurementclassificationecosystem management |
| spellingShingle | Edurne Ibarrola-Ulzurrun Javier Marcello Consuelo Gonzalo-Martin Assessment of Component Selection Strategies in Hyperspectral Imagery Entropy remote sensing hyperspectral sensor feature-extraction texture measurement classification ecosystem management |
| title | Assessment of Component Selection Strategies in Hyperspectral Imagery |
| title_full | Assessment of Component Selection Strategies in Hyperspectral Imagery |
| title_fullStr | Assessment of Component Selection Strategies in Hyperspectral Imagery |
| title_full_unstemmed | Assessment of Component Selection Strategies in Hyperspectral Imagery |
| title_short | Assessment of Component Selection Strategies in Hyperspectral Imagery |
| title_sort | assessment of component selection strategies in hyperspectral imagery |
| topic | remote sensing hyperspectral sensor feature-extraction texture measurement classification ecosystem management |
| url | https://www.mdpi.com/1099-4300/19/12/666 |
| work_keys_str_mv | AT edurneibarrolaulzurrun assessmentofcomponentselectionstrategiesinhyperspectralimagery AT javiermarcello assessmentofcomponentselectionstrategiesinhyperspectralimagery AT consuelogonzalomartin assessmentofcomponentselectionstrategiesinhyperspectralimagery |