Research on segmentation method of hyperspectral remote sensing images based on probabilistic neural networks

Gang Liu; Xingjian Liu

doi:10.1117/12.815625

29 December 2008 Research on segmentation method of hyperspectral remote sensing images based on probabilistic neural networks

Gang Liu, Xingjian Liu

Author Affiliations +

Proceedings Volume 7285, International Conference on Earth Observation Data Processing and Analysis (ICEODPA); 72851J (2008) https://doi.org/10.1117/12.815625
Event: International Conference on Earth Observation Data Processing and Analysis, 2008, Wuhan, China

Abstract

Image segmentation is essential for information extraction from remote sensing image, but it remains the lacks of a general mathematical theory, object merging for poor object boundary localization, dealing with object fragmentation and sensitivity of current procedures to noise. This paper focuses on hyper-spectral image segmentation using probabilistic neural networks (PNN). The methodology, implementation and optimization of a PNN are studied, and a constructed PNN is applied to segment hyper-spectral image. The experience demonstrates main advantage of a PNN that it has quick training and learning, gives a measurement of confidence associated with an output, and has the ability to process large data set. It is concluded that the PNN is superior in image segmentation and the obtained results are satisfied.

Citation Download Citation

Gang Liu and Xingjian Liu "Research on segmentation method of hyperspectral remote sensing images based on probabilistic neural networks", Proc. SPIE 7285, International Conference on Earth Observation Data Processing and Analysis (ICEODPA), 72851J (29 December 2008); https://doi.org/10.1117/12.815625

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available