Hyperspectral Tree Crown Classification Using the Multiple Instance Adaptive Cosine Estimator

Published: July 26th, 2018

Category: Journal Papers


Tree species classification using hyperspectral imagery is a challenging task due to the high spectralsimilarity between species and large intra-species variability. This paper proposes a solution using theMultiple Instance Adaptive Cosine Estimator (MI-ACE) algorithm. MI-ACE estimates a discriminativetarget signature to differentiate between a pair of tree species while accounting for label uncertainty.Additionally, the performance of MI-ACE does not rely on parameter settings that require tuning resultingin a method that is easy to use in application. Results presented are using training and testing dataprovided by a data analysis competition aimed at encouraging the development of methods for extractingecological information through remote sensing obtained through participation in the competition.


“BioRxivLink“Code” “Code”


Zou S, Gader P, Zare A. (2018) Hyperspectral tree crown classification using the multiple instance adaptive cosine estimator. PeerJ Preprints 6:e27052v1 https://doi.org/10.7287/peerj.preprints.27052v1.
Title = {Hyperspectral Tree Crown Classification Using the Multipleinstance Adaptive Cosine Estimator},
Author = {Zou, Z. and Gader, Paul and Zare, A. }
Journal = {PeerJ},
Year = {Under Review},