Abstract
Current multimodal learning in smart feature extraction and classification has reshaped the landscape of remote sensing. The recent developments in smart feature extraction mainly rely on different machine learning and data mining algorithms as powerful classifiers to improve prediction accuracy. This article presents an innovative ensemble learning algorithm for integrated data and classifier fusion via higher order singular value decomposition (IDCF-HOSVD). Based on the fused data, different analytical, semiempirical, and empirical classifiers can be selected and applied to perform information retrieval that can be further synergized via a tensor flow based feature extraction scheme over the classifier space. When preserving core fused image patterns via HOSVD, the final step of IDCF-HOSVD helps rank the contributions from different classifiers via nonlinear correlation information entropy. Practical implementation of the IDCF-HOSVD algorithm was assessed through its application to map the seasonal water quality conditions in Lake Nicaragua, Central America.
Highlights
S ATELLITE remote sensing offers various niches to provide timely monitoring of environmental quality and ecosystem state
This study aims to fill in these technical gaps to advance the capability of multimodal ensemble learning for fused imageries with the aid of integrated higher order singular value decomposition (HOSVD) and nonlinear correlation coefficient (NCC) to advance earth observations
The goal of this study is a seamless integration of data and classifier fusion, resulting in unified neural network learning with machine intelligence, termed “integrated data and classifier fusion with higher order singular value decomposition” (IDCF-HOSVD)
Summary
S ATELLITE remote sensing offers various niches to provide timely monitoring of environmental quality and ecosystem state. Traditional linear regression models (LRMs) have been used on many occasions for feature extraction, and well-known extensions of artificial neural network (ANN) models for feature extraction include morphological shared weight neural networks [5], extreme learning machine (ELM) [6], [7], deep belief nets (DBNs) [8], convolutional neural networks (CNNs) [9], etc. Deep learning, such as DBNs and CNNs, uses multiple-layer architecture to learn representations of input/output data with abstractions from lower to higher levels. The deep learning model was recently used to complete several distinct tasks in different fields, providing better feature extraction capacity [10]–[12]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
