DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Robust Group Subspace Recovery: A New Approach for Multi-Modality Data Fusion

Abstract

Robust Subspace Recovery (RoSuRe) algorithm was recently introduced as a principled and numerically efficient algorithm that unfolds underlying Unions of Subspaces (UoS) structure, present in the data. The union of Subspaces (UoS) is capable of identifying more complex trends in data sets than simple linear models. In this work, we build on and extend RoSuRe to prospect the structure of different data modalities individually. We propose a novel multi-modal data fusion approach based on group sparsity which we refer to as Robust Group Subspace Recovery (RoGSuRe). Relying on a bi-sparsity pursuit paradigm and non-smooth optimization techniques, the introduced framework learns a new joint representation of the time series from different data modalities, respecting an underlying UoS model. We subsequently integrate the obtained structures to form a unified subspace structure. The proposed approach exploits the structural dependencies between the different modalities data to cluster the associated target objects. The resulting fusion of the unlabeled sensors’ data from experiments on audio and magnetic data has shown that our method is competitive with other state of the art subspace clustering methods. The resulting UoS structure is employed to classify newly observed data points, highlighting the abstraction capacity of the proposed method.

Authors:
ORCiD logo [1];  [2]; ORCiD logo [1]; ORCiD logo [3]
  1. North Carolina State Univ., Raleigh, NC (United States)
  2. North Carolina State Univ., Raleigh, NC (United States); Chalmers Univ. of Technology, Gothenburg (Sweden)
  3. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Publication Date:
Research Org.:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1666012
Grant/Contract Number:  
AC05-00OR22725; NA0002576
Resource Type:
Accepted Manuscript
Journal Name:
IEEE Sensors Journal
Additional Journal Information:
Journal Volume: 20; Journal Issue: 20; Journal ID: ISSN 1530-437X
Publisher:
IEEE
Country of Publication:
United States
Language:
English
Subject:
47 OTHER INSTRUMENTATION; Sensor fusion; data integration; data models; sparse matrices; magnetic sensors; sensor phenomena and characterization; sparse learning; unsupervised classification; data fusion; multimodal data

Citation Formats

Ghanem, Sally, Panahi, Ashkan, Krim, Hamid, and Kerekes, Ryan A. Robust Group Subspace Recovery: A New Approach for Multi-Modality Data Fusion. United States: N. p., 2020. Web. doi:10.1109/jsen.2020.2999461.
Ghanem, Sally, Panahi, Ashkan, Krim, Hamid, & Kerekes, Ryan A. Robust Group Subspace Recovery: A New Approach for Multi-Modality Data Fusion. United States. https://doi.org/10.1109/jsen.2020.2999461
Ghanem, Sally, Panahi, Ashkan, Krim, Hamid, and Kerekes, Ryan A. Thu . "Robust Group Subspace Recovery: A New Approach for Multi-Modality Data Fusion". United States. https://doi.org/10.1109/jsen.2020.2999461. https://www.osti.gov/servlets/purl/1666012.
@article{osti_1666012,
title = {Robust Group Subspace Recovery: A New Approach for Multi-Modality Data Fusion},
author = {Ghanem, Sally and Panahi, Ashkan and Krim, Hamid and Kerekes, Ryan A.},
abstractNote = {Robust Subspace Recovery (RoSuRe) algorithm was recently introduced as a principled and numerically efficient algorithm that unfolds underlying Unions of Subspaces (UoS) structure, present in the data. The union of Subspaces (UoS) is capable of identifying more complex trends in data sets than simple linear models. In this work, we build on and extend RoSuRe to prospect the structure of different data modalities individually. We propose a novel multi-modal data fusion approach based on group sparsity which we refer to as Robust Group Subspace Recovery (RoGSuRe). Relying on a bi-sparsity pursuit paradigm and non-smooth optimization techniques, the introduced framework learns a new joint representation of the time series from different data modalities, respecting an underlying UoS model. We subsequently integrate the obtained structures to form a unified subspace structure. The proposed approach exploits the structural dependencies between the different modalities data to cluster the associated target objects. The resulting fusion of the unlabeled sensors’ data from experiments on audio and magnetic data has shown that our method is competitive with other state of the art subspace clustering methods. The resulting UoS structure is employed to classify newly observed data points, highlighting the abstraction capacity of the proposed method.},
doi = {10.1109/jsen.2020.2999461},
journal = {IEEE Sensors Journal},
number = 20,
volume = 20,
place = {United States},
year = {Thu Oct 15 00:00:00 EDT 2020},
month = {Thu Oct 15 00:00:00 EDT 2020}
}

Works referenced in this record:

Two decades of array signal processing research: the parametric approach
journal, July 1996

  • Krim, H.; Viberg, M.
  • IEEE Signal Processing Magazine, Vol. 13, Issue 4
  • DOI: 10.1109/79.526899

Sparse Subspace Clustering: Algorithm, Theory, and Applications
preprint, January 2012


Distributed classification of acoustic targets in wireless audio-sensor networks
journal, September 2008


Bi-sparsity pursuit: A paradigm for robust subspace recovery
journal, November 2018


Clustering disjoint subspaces via sparse representation
conference, March 2010

  • Elhamifar, Ehsan; Vidal, Rene
  • 2010 IEEE International Conference on Acoustics, Speech and Signal Processing
  • DOI: 10.1109/ICASSP.2010.5495317

Sparse subspace clustering
conference, June 2009

  • Elhamifar, Ehsan; Vidal, Rene
  • 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2009 IEEE Conference on Computer Vision and Pattern Recognition
  • DOI: 10.1109/CVPR.2009.5206547

Sparse Subspace Clustering: Algorithm, Theory, and Applications
journal, November 2013

  • Elhamifar, E.; Vidal, R.
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, Issue 11
  • DOI: 10.1109/TPAMI.2013.57

Multimodal sparse and low-rank subspace clustering
journal, January 2018


Low-Rank Tensor Constrained Multiview Subspace Clustering
conference, December 2015

  • Zhang, Changqing; Fu, Huazhu; Liu, Si
  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • DOI: 10.1109/ICCV.2015.185

Robust Recovery of Subspace Structures by Low-Rank Representation
journal, January 2013

  • Liu, Guangcan; Lin, Zhouchen; Yan, Shuicheng
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, Issue 1
  • DOI: 10.1109/TPAMI.2012.88

A geometric analysis of subspace clustering with outliers
journal, August 2012

  • Soltanolkotabi, Mahdi; Candés, Emmanuel J.
  • The Annals of Statistics, Vol. 40, Issue 4
  • DOI: 10.1214/12-AOS1034

Enhanced Detectability of Community Structure in Multilayer Networks through Layer Aggregation
journal, June 2016


Information Subspace-Based Fusion for Vehicle Classification
conference, September 2018


Multisensor data fusion: A review of the state-of-the-art
journal, January 2013


Multi-task low-rank affinity pursuit for image segmentation
conference, November 2011


On the Role of Sparse and Redundant Representations in Image Processing
journal, June 2010


Robust principal component analysis?
journal, May 2011


Vehicle detection and localization using Unattended Ground Magnetometer Sensors
conference, July 2010


RFTraffic: a study of passive traffic awareness using emitted RF noise from the vehicles
journal, January 2012

  • Ding, Yong; Banitalebi, Behnam; Miyaki, Takashi
  • EURASIP Journal on Wireless Communications and Networking, Vol. 2012, Issue 1
  • DOI: 10.1186/1687-1499-2012-8

Two decades of array signal processing research: the parametric approach
journal, July 1996

  • Krim, H.; Viberg, M.
  • IEEE Signal Processing Magazine, Vol. 13, Issue 4
  • DOI: 10.1109/79.526899

Critiques on some combination rules for probability theory based on optimization techniques
conference, July 2007

  • Florea, Mihai Cristian; Bosse, Eloi
  • 2007 10th International Conference on Information Fusion
  • DOI: 10.1109/ICIF.2007.4408062

Convex optimization approach to identify fusion for multisensor target tracking
journal, May 2001

  • Lingjie Li, ; Wong, K. M.
  • IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, Vol. 31, Issue 3
  • DOI: 10.1109/3468.925656

An introduction to multisensor data fusion
journal, January 1997

  • Hall, D. L.; Llinas, J.
  • Proceedings of the IEEE, Vol. 85, Issue 1
  • DOI: 10.1109/5.554205

Isoradial Bodies
journal, August 2004

  • Brandenberg, Ren�; Dattasharma, Abhi; Gritzmann, Peter
  • Discrete & Computational Geometry, Vol. 32, Issue 4
  • DOI: 10.1007/s00454-004-1132-4

Multimodal Data Fusion: An Overview of Methods, Challenges, and Prospects
journal, September 2015


Vehicle Classification and Identification Using Multi-Modal Sensing and Signal Learning
conference, June 2017


On the use of instantaneous and transitional spectral information in speaker recognition
journal, June 1988

  • Soong, F. K.; Rosenberg, A. E.
  • IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. 36, Issue 6
  • DOI: 10.1109/29.1598

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

Prediction Intervals to Account for Uncertainties in Travel Time Prediction
journal, June 2011

  • Khosravi, Abbas; Mazloumi, Ehsan; Nahavandi, Saeid
  • IEEE Transactions on Intelligent Transportation Systems, Vol. 12, Issue 2
  • DOI: 10.1109/TITS.2011.2106209

The use of RFID for traffic management
journal, January 2008

  • Kathawala, Yunus A.; Tueck, Benjamin
  • International Journal of Technology, Policy and Management, Vol. 8, Issue 2
  • DOI: 10.1504/IJTPM.2008.017215

Vehicle re-identification using wireless magnetic sensors: Algorithm revision, modifications and performance analysis
conference, July 2011

  • Sanchez, Rene O.; Flores, Christopher; Horowitz, Roberto
  • 2011 IEEE International Conference on Vehicular Electronics and Safety (ICVES 2011), Proceedings of 2011 IEEE International Conference on Vehicular Electronics and Safety
  • DOI: 10.1109/ICVES.2011.5983819

Wireless magnetic sensors for traffic surveillance
journal, June 2008

  • Haoui, Amine; Kavaler, Robert; Varaiya, Pravin
  • Transportation Research Part C: Emerging Technologies, Vol. 16, Issue 3
  • DOI: 10.1016/j.trc.2007.10.004

Kernel sparse subspace clustering
conference, October 2014


Multiscale Hybrid Linear Models for Lossy Image Representation
journal, December 2006

  • Hong, Wei; Wright, John; Huang, Kun
  • IEEE Transactions on Image Processing, Vol. 15, Issue 12
  • DOI: 10.1109/TIP.2006.882016

Analysis Dictionary Learning Based Classification: Structure for Robustness
journal, December 2019

  • Tang, Wen; Panahi, Ashkan; Krim, Hamid
  • IEEE Transactions on Image Processing, Vol. 28, Issue 12
  • DOI: 10.1109/TIP.2019.2919409

Bi-sparsity pursuit: A paradigm for robust subspace recovery
journal, November 2018


A Note on the Use of Principal Components in Regression
journal, January 1982

  • Jolliffe, Ian T.
  • Applied Statistics, Vol. 31, Issue 3
  • DOI: 10.2307/2348005

Speaker Verification Using Adapted Gaussian Mixture Models
journal, January 2000

  • Reynolds, Douglas A.; Quatieri, Thomas F.; Dunn, Robert B.
  • Digital Signal Processing, Vol. 10, Issue 1-3
  • DOI: 10.1006/dspr.1999.0361

Improved adaptive Gaussian mixture model for background subtraction
conference, January 2004

  • Zivkovic, Z.
  • Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004.
  • DOI: 10.1109/ICPR.2004.1333992

Distributed classification of acoustic targets in wireless audio-sensor networks
journal, September 2008


Vehicle re-identification with a single magnetic sensor
conference, May 2012

  • Charbonnier, Sylvie; Pitton, Anne-Cecile; Vassilev, Andrea
  • 2012 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), 2012 IEEE International Instrumentation and Measurement Technology Conference Proceedings
  • DOI: 10.1109/I2MTC.2012.6229117