Skip to main content

An improved classification method for hyperspectral data based on spectral and morphological information

Buy Article:

$63.00 plus tax (Refund Policy)

Abstract:

Combining spectral and morphological information to classify hyperspectral data offers a considerable advantage over methods based solely on spectral or morphological data. Previously, a classification method was proposed that concatenated extended morphological profile (EMP) and spectral information into one feature vector for each pixel of an image. Although this method has the merit of simultaneously using spectral and morphological information, it runs the risk of generating new spectral constituents (not present in the original image). In this letter, an improved classification method based on fusing extended multivariate morphological profile (EMMP) and spectral information was proposed. A new vector ordering method based on spectral purity-based criterion was adopted to overcome the problem of generating new constituents in EMP. A feature selection algorithm was employed to improve the efficiency of EMMP. Experiments were carried out on a hyperspectral data set collected by NASA's Airborne Visible-Infrared Imaging Spectrometer sensor. Experimental results showed that EMMP was effective at describing morphological information in hyperspectral data, and that this letter's method was superior to the previous method in terms of classification accuracy but inferior to the previous method in terms of time consumption.

Document Type: Research Article

DOI: https://doi.org/10.1080/01431161.2010.510488

Affiliations: 1: Center for Earth Observation and Digital Earth, Chinese Academy of Sciences, Beijing, PR China,Graduate University of Chinese Academy of Sciences, Beijing, PR China 2: Center for Earth Observation and Digital Earth, Chinese Academy of Sciences, Beijing, PR China 3: Department of Geography and Geoinformation Science, George Mason University, Fairfax, Virginia, USA

Publication date: 2011-05-01

More about this publication?
  • Access Key
  • Free content
  • Partial Free content
  • New content
  • Open access content
  • Partial Open access content
  • Subscribed content
  • Partial Subscribed content
  • Free trial content
Cookie Policy
X
Cookie Policy
Ingenta Connect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more