Functions of Multiple Instances for Learning Target Signatures
Authors:
- 11 March 2015
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 53 (8) , 4670-4686
Abstract
The functions of multiple instances (FUMI) approach for learning target and nontarget signatures is introduced. FUMI is a generalization of the multiple-instance 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 (cFUMI) and extended cFUMI (eFUMI), are presented and applied to the problem of hyperspectral unmixing and target detection. cFUMI 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 eFUMI algorithm extends the cFUMI 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 eFUMI) 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.Keywords
Funding Information
- National Science Foundation (IIS-1350078-CAREER)
This publication has 44 references indexed in Scilit:
- A Robust Nonlinear Hyperspectral Anomaly Detection ApproachIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014
- Target detection based on a dynamic subspacePattern Recognition, 2014
- Spatial-Spectral Information Based Abundance-Constrained Endmember Extraction MethodsIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2013
- Subspace Matching Pursuit for Sparse Unmixing of Hyperspectral DataIEEE Transactions on Geoscience and Remote Sensing, 2013
- Sparse Hyperspectral Unmixing Based on Constrained lp - l2 OptimizationIEEE Geoscience and Remote Sensing Letters, 2013
- Piece-wise convex spatial-spectral unmixing of hyperspectral imagery using possibilistic and fuzzy clustering2011
- Multiclass Relevance Vector Machines: Sparsity and AccuracyIEEE Transactions on Neural Networks, 2010
- Spatially-smooth piece-wise convex endmember detection2010
- Integration of spatial–spectral information for the improved extraction of endmembersRemote Sensing of Environment, 2007
- Spatial/spectral endmember extraction by multidimensional morphological operationsIEEE Transactions on Geoscience and Remote Sensing, 2002