Ensemble learning of satellite remote sensing images via integrating deep and fast learning algorithms for water quality monitoring (Conference Presentation)

Ni-Bin Chang; Zhibin Sun; Wei Gao

doi:10.1117/12.2319432

18 September 2018 Ensemble learning of satellite remote sensing images via integrating deep and fast learning algorithms for water quality monitoring (Conference Presentation)

Ni-Bin Chang, Zhibin Sun, Wei Gao

Author Affiliations +

Proceedings Volume 10767, Remote Sensing and Modeling of Ecosystems for Sustainability XV; 107670D (2018) https://doi.org/10.1117/12.2319432
Event: SPIE Optical Engineering + Applications, 2018, San Diego, California, United States

Abstract

Previous remote sensing studies of intelligent feature extraction led to the successful image fusion, merging, and cloudy pixel reconstruction destined for the spatiotemporal change detection. Based on fused satellite images with better spatial and temporal resolution, this study explores a thorough comparative analysis in terms of feature extraction capability of deep learning, regular learning, fast learning, and ensemble learning relative to some traditional feature extraction algorithms (2-band and linear regression models). In specific, this study aims to evaluate the systematic influences of fast and deep learning models with potential to create a new ensemble learning tool for better feature extraction based on fused remote sensing images. In ensemble learning step, the whole ground-truth dataset is fed into the selected ensemble learning algorithm (i.e., a classifier fusion algorithm) with the aid of singular value decomposition to create an integrative tool. Practical implementation was assessed by a case study of water quality monitoring over dry and wet seasons in Lake Nicaragua, Central America. Both deep and fast learning algorithms outperform the regular learning algorithm with a single layer forward network and ensemble learning may take advantage of merits of deep, fast, and regular learning algorithms. Final water quality assessment was generated based on the integrative algorithm of the two with bio-optical models for eutrophication assessment in Lake Nicaragua. Although deep learning has better results in validation and the ensemble learning model aggregates different types of strength from all models based on all limited ground-truth samples.

Conference Presentation

Citation Download Citation

Ni-Bin Chang, Zhibin Sun, and Wei Gao "Ensemble learning of satellite remote sensing images via integrating deep and fast learning algorithms for water quality monitoring (Conference Presentation)", Proc. SPIE 10767, Remote Sensing and Modeling of Ecosystems for Sustainability XV, 107670D (18 September 2018); https://doi.org/10.1117/12.2319432

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

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
PRESENTATION

WATCH
PRESENTATION SAVE TO MY LIBRARY

GET CITATION

KEYWORDS

Remote sensing

Earth observing sensors

Image fusion

Satellite imaging

Satellites

Feature extraction

Reconstruction algorithms

Show All Keywords

Keywords/Phrases

Search In:

Publication Years