skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A statistical approach to combining multisource information in one-class classifiers

Abstract

A new method is introduced in this paper for combining information from multiple sources to support one-class classification. The contributing sources may represent measurements taken by different sensors of the same physical entity, repeated measurements by a single sensor, or numerous features computed from a single measured image or signal. The approach utilizes the theory of statistical hypothesis testing, and applies Fisher's technique for combining p-values, modified to handle nonindependent sources. Classifier outputs take the form of fused p-values, which may be used to gauge the consistency of unknown entities with one or more class hypotheses. The approach enables rigorous assessment of classification uncertainties, and allows for traceability of classifier decisions back to the constituent sources, both of which are important for high-consequence decision support. Application of the technique is illustrated in two challenge problems, one for skin segmentation and the other for terrain labeling. Finally, the method is seen to be particularly effective for relatively small training samples.

Authors:
ORCiD logo [1];  [1];  [1];  [1];  [1]
  1. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Publication Date:
Research Org.:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA); SNL Laboratory Directed Research and Development (LDRD) program
OSTI Identifier:
1399493
Alternate Identifier(s):
OSTI ID: 1400859
Report Number(s):
SAND2017-2026J
Journal ID: ISSN 1932-1864; 651299
Grant/Contract Number:  
AC04-94AL85000
Resource Type:
Accepted Manuscript
Journal Name:
Statistical Analysis and Data Mining
Additional Journal Information:
Journal Volume: 10; Journal Issue: 4; Journal ID: ISSN 1932-1864
Publisher:
Wiley
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; classification; dependent p-values; Fisher's combination method; gamma distribution; image segmentation; multisource fusion

Citation Formats

Simonson, Katherine M., Derek West, R., Hansen, Ross L., LaBruyere, Thomas E., and Van Benthem, Mark H. A statistical approach to combining multisource information in one-class classifiers. United States: N. p., 2017. Web. doi:10.1002/sam.11342.
Simonson, Katherine M., Derek West, R., Hansen, Ross L., LaBruyere, Thomas E., & Van Benthem, Mark H. A statistical approach to combining multisource information in one-class classifiers. United States. doi:10.1002/sam.11342.
Simonson, Katherine M., Derek West, R., Hansen, Ross L., LaBruyere, Thomas E., and Van Benthem, Mark H. Thu . "A statistical approach to combining multisource information in one-class classifiers". United States. doi:10.1002/sam.11342. https://www.osti.gov/servlets/purl/1399493.
@article{osti_1399493,
title = {A statistical approach to combining multisource information in one-class classifiers},
author = {Simonson, Katherine M. and Derek West, R. and Hansen, Ross L. and LaBruyere, Thomas E. and Van Benthem, Mark H.},
abstractNote = {A new method is introduced in this paper for combining information from multiple sources to support one-class classification. The contributing sources may represent measurements taken by different sensors of the same physical entity, repeated measurements by a single sensor, or numerous features computed from a single measured image or signal. The approach utilizes the theory of statistical hypothesis testing, and applies Fisher's technique for combining p-values, modified to handle nonindependent sources. Classifier outputs take the form of fused p-values, which may be used to gauge the consistency of unknown entities with one or more class hypotheses. The approach enables rigorous assessment of classification uncertainties, and allows for traceability of classifier decisions back to the constituent sources, both of which are important for high-consequence decision support. Application of the technique is illustrated in two challenge problems, one for skin segmentation and the other for terrain labeling. Finally, the method is seen to be particularly effective for relatively small training samples.},
doi = {10.1002/sam.11342},
journal = {Statistical Analysis and Data Mining},
number = 4,
volume = 10,
place = {United States},
year = {2017},
month = {6}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 1 work
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

The Optics of Human Skin
journal, July 1981


Active and dynamic information fusion for facial expression understanding from image sequences
journal, May 2005

  • Yongmian Zhang,
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27, Issue 5
  • DOI: 10.1109/TPAMI.2005.93

Growing a multi-class classifier with a reject option
journal, July 2008


General Four-Component Scattering Power Decomposition With Unitary Transformation of Coherency Matrix
journal, May 2013

  • Singh, Gulab; Yamaguchi, Yoshio; Park, Sang-Eun
  • IEEE Transactions on Geoscience and Remote Sensing, Vol. 51, Issue 5
  • DOI: 10.1109/TGRS.2012.2212446

Log-polar wavelet energy signatures for rotation and scale invariant texture classification
journal, May 2003


Classification and Extraction of Spatial Features in Urban Areas Using High-Resolution Multispectral Imagery
journal, April 2007

  • Huang, Xin; Zhang, Liangpei; Li, Pingxiang
  • IEEE Geoscience and Remote Sensing Letters, Vol. 4, Issue 2
  • DOI: 10.1109/LGRS.2006.890540

Some bivariate uniform distributions
journal, January 1980


Polarimetric SAR interferometry
journal, January 1998

  • Cloude, S. R.; Papathanassiou, K. P.
  • IEEE Transactions on Geoscience and Remote Sensing, Vol. 36, Issue 5
  • DOI: 10.1109/36.718859

Learning Hierarchical Features for Scene Labeling
journal, August 2013

  • Farabet, Clement; Couprie, Camille; Najman, Laurent
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, Issue 8
  • DOI: 10.1109/TPAMI.2012.231

Robust Real-Time Unusual Event Detection using Multiple Fixed-Location Monitors
journal, March 2008

  • Adam, Amit; Rivlin, Ehud; Shimshoni, Ilan
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 30, Issue 3
  • DOI: 10.1109/TPAMI.2007.70825

A survey on feature selection methods
journal, January 2014


A systematic comparison of methods for combining p-values from independent tests
journal, October 2004


Soft computing methods applied to combination of one-class classifiers
journal, January 2012


The strength of weak learnability
journal, June 1990


Estimating the Support of a High-Dimensional Distribution
journal, July 2001


Evolving feature selection
journal, November 2005

  • Liu, H.; Dougherty, E. R.; Dy, J. G.
  • IEEE Intelligent Systems, Vol. 20, Issue 6
  • DOI: 10.1109/MIS.2005.105

Multiple classifiers applied to multisource remote sensing data
journal, January 2002

  • Briem, G. J.; Benediktsson, J. A.; Sveinsson, J. R.
  • IEEE Transactions on Geoscience and Remote Sensing, Vol. 40, Issue 10
  • DOI: 10.1109/TGRS.2002.802476

Wavelet packet feature extraction for vibration monitoring
journal, June 2000

  • Yen, G. G.; Lin, K. -C.
  • IEEE Transactions on Industrial Electronics, Vol. 47, Issue 3
  • DOI: 10.1109/41.847906

Asymptotic Optimality of Fisher's Method of Combining Independent Tests II
journal, March 1973


A review of novelty detection
journal, June 2014


A Functional Data—Analytic Approach to Signal Discrimination
journal, February 2001


Distribution of Fisher's combination statistic when the tests are dependent
journal, January 2010


ECG Beats Classification Using Mixture of Features
journal, January 2014

  • Das, Manab Kumar; Ari, Samit
  • International Scholarly Research Notices, Vol. 2014
  • DOI: 10.1155/2014/178436

Statistical pattern recognition: a review
journal, January 2000

  • Jain, A. K.; Duin, P. W.
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, Issue 1
  • DOI: 10.1109/34.824819

Expert guided natural language processing using one-class classification
journal, June 2015

  • Joffe, Erel; Pettigrew, Emily J.; Herskovic, Jorge R.
  • Journal of the American Medical Informatics Association, Vol. 22, Issue 5
  • DOI: 10.1093/jamia/ocv010

A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting
journal, August 1997

  • Freund, Yoav; Schapire, Robert E.
  • Journal of Computer and System Sciences, Vol. 55, Issue 1
  • DOI: 10.1006/jcss.1997.1504

Support-vector networks
journal, September 1995

  • Cortes, Corinna; Vapnik, Vladimir
  • Machine Learning, Vol. 20, Issue 3
  • DOI: 10.1007/BF00994018

SLIC Superpixels Compared to State-of-the-Art Superpixel Methods
journal, November 2012

  • Achanta, R.; Shaji, A.; Smith, K.
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 34, Issue 11, p. 2274-2282
  • DOI: 10.1109/TPAMI.2012.120

An Empirical Comparison of Expert-Derived and Data-Derived Classification Trees
journal, January 1996


Combining dependent P-values
journal, November 2002


Bagging predictors
journal, August 1996


A Fast Learning Algorithm for Deep Belief Nets
journal, July 2006


Combination of multiple classifiers using local accuracy estimates
journal, April 1997

  • Woods, K.; Kegelmeyer, W. P.; Bowyer, K.
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 19, Issue 4
  • DOI: 10.1109/34.588027

From colour to tissue histology: Physics-based interpretation of images of pigmented skin lesions
journal, December 2003


A New Coherency Formalism for Change Detection and Phenomenology in SAR Imagery: A Field Approach
journal, July 2009


Methods of combining multiple classifiers and their applications to handwriting recognition
journal, January 1992

  • Xu, L.; Krzyzak, A.; Suen, C. Y.
  • IEEE Transactions on Systems, Man, and Cybernetics, Vol. 22, Issue 3
  • DOI: 10.1109/21.155943

One-class classification based authentication of peanut oils by fatty acid profiles
journal, January 2015

  • Zhang, Liangxiao; Li, Peiwu; Sun, Xiaoman
  • RSC Advances, Vol. 5, Issue 103
  • DOI: 10.1039/C5RA07329D

An overview of text-independent speaker recognition: From features to supervectors
journal, January 2010


On the Combination of Independent Test Statistics
journal, June 1967

  • van Zwet, W. R.; Oosterhoff, J.
  • The Annals of Mathematical Statistics, Vol. 38, Issue 3
  • DOI: 10.1214/aoms/1177698861

Effective dimension reduction methods for tumor classification using gene expression data
journal, March 2003


400: A Method for Combining Non-Independent, One-Sided Tests of Significance
journal, December 1975


The FERET evaluation methodology for face-recognition algorithms
conference, January 1997

  • Phillips, P. J.; Rauss, P.
  • Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition
  • DOI: 10.1109/CVPR.1997.609311

    Works referencing / citing this record:

    Dataset for: A Statistical Approach to Combining Multi-Source Information in One-Class Classifiers [Supplemental Data]
    dataset, June 2017

    • Simonson, Katherine; West, R. Derek; Hansen, Ross
    • figshare-Supplementary information for journal article at DOI: 10.1002/sam.11342, 10 TXT files
    • DOI: 10.6084/m9.figshare.c.3682954