Semi-Supervised Based Hyperspectral Imagery Classification

Hyperspectral imagery classification is a challenging problem. Wherein, the high number of spectral channels and the high cost of true sample labeling greatly reduce the classification precision. In this paper, we proposed a semi-supervised method, which combine linear discriminant analysis and mani...

Popoln opis

Shranjeno v:
Bibliografske podrobnosti
Main Authors: Zhijun Zheng, Yanbin Peng
Format: Artigo
Jezik:Inglês
Izdano: IFSA Publishing, S.L. 2013-09-01
Serija:Sensors & Transducers
Teme:
Hyperspectral imagery
Classification
Semi-supervised
Linear discriminant analysis
Manifold learning.
Online dostop:http://www.sensorsportal.com/HTML/DIGEST/september_2013/P_1360.pdf
Oznake: Označite
Brez oznak, prvi označite!

Podobne knjige/članki