干旱区生态环境知识资源中心(Arid): An Architecture to Classify Desertification Areas using Hyperspectral Images and the Optimum Path Forest Algorithm

Arid

	An Architecture to Classify Desertification Areas using Hyperspectral Images and the Optimum Path Forest Algorithm
	Macedo, Marcia R. O. B. C.; Times, Valeria C.; Cavalcanti, George D. C.; Kohlman Rabbani, Emilia Rahnemay
通讯作者	Macedo, MROBC
来源期刊	ELECTRONIC JOURNAL OF GEOTECHNICAL ENGINEERING
ISSN	1089-3032
出版年	2016
卷号	21 期号:5 页码:1881-1895
英文摘要	The desertification is the process of degradation of the lands of arid regions, semiarid and dry sub humid, resulting of different factors, among them, the climatic variation and human activity. Inappropriate land use and soil mismanagement are the most important anthropic causes of desertification. Traditionally, multispectral sensors with a small number of bands have been used in remote sensing to discriminate most classes that occur in natural scenes such as vegetation, water bodies, soils and urban areas. But that is not sufficient when we try to discriminate many ground cover classes that do not have simple, uniquely identifiable reflectance spectra. The progress in remote sensing technology over the recent years has lead to the launch of hyperspectral remote sensing systems. The present study is exploring the potential of Hyperion hyperspectral imagery combined with Optimum Forest Path (OPF) algorithm for supervised classification of areas affected by desertification process and compares the efficacy between the OPF and SVM classifiers when applied to these areas. Validation of the land cover thematic maps was performed based on the confusion matrix analysis using for consistency the same set of validation points. Both classifiers produced generally reasonable results with the OPF however significantly outperforming the SVM in overall classification accuracy. The higher classification accuracy by OPF was attributed principally to the ability to identify a better distinction between the regions of degraded areas (DA, DOC and DP) and preserved areas (PGP and PDP).
英文关键词	remote sensing image processing desertification
类型	Article
语种	英语
收录类别	ESCI
WOS记录号	WOS:000459504800015
WOS类目	Engineering, Geological
WOS研究方向	Engineering
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/332048
作者单位	[Macedo, Marcia R. O. B. C.; Kohlman Rabbani, Emilia Rahnemay] Univ Pernambuco, Polytech Sch Pernambuco, Recife, PE, Brazil; [Times, Valeria C.; Cavalcanti, George D. C.] Univ Fed Pernambuco, Ctr Informat, Recife, PE, Brazil
推荐引用方式 GB/T 7714	Macedo, Marcia R. O. B. C.,Times, Valeria C.,Cavalcanti, George D. C.,et al. An Architecture to Classify Desertification Areas using Hyperspectral Images and the Optimum Path Forest Algorithm[J],2016,21(5):1881-1895.
APA	Macedo, Marcia R. O. B. C.,Times, Valeria C.,Cavalcanti, George D. C.,&Kohlman Rabbani, Emilia Rahnemay.(2016).An Architecture to Classify Desertification Areas using Hyperspectral Images and the Optimum Path Forest Algorithm.ELECTRONIC JOURNAL OF GEOTECHNICAL ENGINEERING,21(5),1881-1895.
MLA	Macedo, Marcia R. O. B. C.,et al."An Architecture to Classify Desertification Areas using Hyperspectral Images and the Optimum Path Forest Algorithm".ELECTRONIC JOURNAL OF GEOTECHNICAL ENGINEERING 21.5(2016):1881-1895.