Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information

An incremental classification algorithm INC_SPEC_MP<sub>ext</sub> was proposed for hyperspectral remote sensing images based on spectral and spatial information. The spatial information was extracted by building morphological profiles based on several principle components of hyperspectra...

Full description

Bibliographic Details
Main Authors: WANG Junshu, JIANG Nan, ZHANG Guoming, LI Yang, LV Heng
Format: Article
Language:zho
Published: Surveying and Mapping Press 2015-09-01
Series:Acta Geodaetica et Cartographica Sinica
Subjects:
Online Access:http://html.rhhz.net/CHXB/html/2015-9-1003.htm
id doaj-571547bd2c3946689d396afe3cf64379
record_format Article
spelling doaj-571547bd2c3946689d396afe3cf643792020-11-24T23:30:50ZzhoSurveying and Mapping PressActa Geodaetica et Cartographica Sinica1001-15951001-15952015-09-014491003101310.11947/j.AGCS.2015.2014038820150909Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial InformationWANG Junshu0JIANG Nan1ZHANG Guoming2LI Yang3LV Heng4Key Laboratory for Virtual Geographic Environment, Ministry of Education, Nanjing Normal University, Nanjing 210023, China;Key Laboratory for Virtual Geographic Environment, Ministry of Education, Nanjing Normal University, Nanjing 210023, China;Center of Health Statistics and Information of Jiangsu Province, Nanjing 210008, ChinaKey Laboratory for Virtual Geographic Environment, Ministry of Education, Nanjing Normal University, Nanjing 210023, China;Key Laboratory for Virtual Geographic Environment, Ministry of Education, Nanjing Normal University, Nanjing 210023, China;An incremental classification algorithm INC_SPEC_MP<sub>ext</sub> was proposed for hyperspectral remote sensing images based on spectral and spatial information. The spatial information was extracted by building morphological profiles based on several principle components of hyperspectral image. The morphological profiles were combined together in extended morphological profiles (MP<sub>ext</sub>). Combine spectral and MP<sub>ext</sub> to enrich knowledge and utilize the useful information of unlabeled data at the most extent to optimize the classifier. Pick out high confidence data and add to training set, then retrain the classifier with augmented training set to predict the rest samples. The process was performed iteratively. The proposed algorithm was tested on AVIRIS Indian Pines and Hyperion EO-1 Botswana data, which take on different covers, and experimental results show low classification cost and significant improvements in terms of accuracies and Kappa coefficient under limited training samples compared with the classification results based on spectral, MP<sub>ext</sub> and the combination of sepctral and MP<sub>ext</sub>.http://html.rhhz.net/CHXB/html/2015-9-1003.htmhyperspectral remote sensing imagemorphologyspatial informationspectral informationincremental classification
collection DOAJ
language zho
format Article
sources DOAJ
author WANG Junshu
JIANG Nan
ZHANG Guoming
LI Yang
LV Heng
spellingShingle WANG Junshu
JIANG Nan
ZHANG Guoming
LI Yang
LV Heng
Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information
Acta Geodaetica et Cartographica Sinica
hyperspectral remote sensing image
morphology
spatial information
spectral information
incremental classification
author_facet WANG Junshu
JIANG Nan
ZHANG Guoming
LI Yang
LV Heng
author_sort WANG Junshu
title Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information
title_short Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information
title_full Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information
title_fullStr Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information
title_full_unstemmed Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information
title_sort incremental classification algorithm of hyperspectral remote sensing images based on spectral-spatial information
publisher Surveying and Mapping Press
series Acta Geodaetica et Cartographica Sinica
issn 1001-1595
1001-1595
publishDate 2015-09-01
description An incremental classification algorithm INC_SPEC_MP<sub>ext</sub> was proposed for hyperspectral remote sensing images based on spectral and spatial information. The spatial information was extracted by building morphological profiles based on several principle components of hyperspectral image. The morphological profiles were combined together in extended morphological profiles (MP<sub>ext</sub>). Combine spectral and MP<sub>ext</sub> to enrich knowledge and utilize the useful information of unlabeled data at the most extent to optimize the classifier. Pick out high confidence data and add to training set, then retrain the classifier with augmented training set to predict the rest samples. The process was performed iteratively. The proposed algorithm was tested on AVIRIS Indian Pines and Hyperion EO-1 Botswana data, which take on different covers, and experimental results show low classification cost and significant improvements in terms of accuracies and Kappa coefficient under limited training samples compared with the classification results based on spectral, MP<sub>ext</sub> and the combination of sepctral and MP<sub>ext</sub>.
topic hyperspectral remote sensing image
morphology
spatial information
spectral information
incremental classification
url http://html.rhhz.net/CHXB/html/2015-9-1003.htm
work_keys_str_mv AT wangjunshu incrementalclassificationalgorithmofhyperspectralremotesensingimagesbasedonspectralspatialinformation
AT jiangnan incrementalclassificationalgorithmofhyperspectralremotesensingimagesbasedonspectralspatialinformation
AT zhangguoming incrementalclassificationalgorithmofhyperspectralremotesensingimagesbasedonspectralspatialinformation
AT liyang incrementalclassificationalgorithmofhyperspectralremotesensingimagesbasedonspectralspatialinformation
AT lvheng incrementalclassificationalgorithmofhyperspectralremotesensingimagesbasedonspectralspatialinformation
_version_ 1725540134871367680