Functions of Multiple Instances for Learning Target Signatures


Abstract
:

The functions of multiple instances (FUMI) approach for learning target and nontarget signatures is introduced. FUMI is a generalization of the multipleinstance learning (MIL) approach for supervised learning. FUMI differs significantly from standard MIL and supervised learning approaches because only data points which are functions of class concepts/signatures are available. In particular, this paper addresses the problem in which data points are convex combinations of target and nontarget signatures. Two algorithms, convex FUMI ( c FUMI) and extended c FUMI ( e FUMI), are presented and applied to the problem of hyperspectral unmixing and target detection. c FUMI learns target and nontarget signatures (i.e., target and nontarget endmembers), the number of nontarget signatures, and the proportion of each signature for every data point. The e FUMI algorithm extends the c FUMI to allow for additional “bag level” uncertainty in training labels. For these methods, training data need only binary labels indicating whether a data point (or some spatial area in the case of e FUMI) contains or does not contain some proportion of the target; the specific target proportions for the training data are not needed. After learning the target signature using the binary-labeled training data, target detection can be performed on test data. Results for subpixel target detection on simulated and real airborne hyperspectral data are shown.

Links:

Document

IEEE Abstract

Google Scholar

Reference:

Plain Text:

C. Jiao, A. Zare,  “Functions of Multiple Instances for Learning Target Signatures,”  IEEE Trans. Geosci. Remote Sens., vol. 53, no. 8, pp. 4670-4686, Aug. 2015

BibTex:

@Article{Jiao2015Functions,
Title = {Functions of Multiple Instances for Learning Target Signatures},
Author = {Jiao, Changzhe and Zare, Alina},
Journal = {IEEE Trans. Geosci. Remote Sens.},
Year = {2015},
Month = {Aug.},
Number = {8},
Pages = {4670-4686},
Volume = {53}
}

 

 

Categorized as:


Tags