3-D radar image processing methodology for Non-Destructive Testing of aeronautics composite materials and structures

Recently, there has been a significant interest in employing high frequency radar imagery for many industrial oriented and security applications. The quality control of aeronautics composite multi-layered materials and structures through Non-Destructive Testing (NDT) is the main focus of this study. The main motivations for using this technology are: it allows penetration of most non-metal and non-polarized materials, it provides the ability for three-dimensional (3-D) imagery and in-depth information, and the millimeter waves (mmW) pose no health risk to the operator. The main goal of this research is to develop an integrated, semi-automatic and near real-time mode-operated image processing methodology for frequency-modulated continuous-wave (FMCW) millimeter wave images with center frequencies around 100 GHz and 300 GHz. The proposed methodology firstly foresees to process reflectance and transmittance 3-D imagery by extracting areas of interest and object's boundaries at two operated frequencies. Then, the detected areas are subjected to a multi-source database and integrated by a decision tree algorithm. The fused information is used to identify defects and in-homogeneities within the objects. Finally, the post-processing phase examines and evaluates the spatial accuracy of the extracted information.

[1]  Alexander A. Nepomnyashchy,et al.  Advances in Sensing with Security Applications , 2006 .

[2]  N. Llombart,et al.  Penetrating 3-D Imaging at 4- and 25-m Range Using a Submillimeter-Wave Radar , 2008, IEEE Transactions on Microwave Theory and Techniques.

[3]  Lucila Ohno-Machado,et al.  Logistic regression and artificial neural network classification models: a methodology review , 2002, J. Biomed. Informatics.

[4]  Arttu Luukanen,et al.  Terahertz active direct detection imagers , 2004, SPIE Defense + Commercial Sensing.

[5]  Jen-Tzung Chien,et al.  A hybrid algorithm for speaker adaptation using MAP transformation and adaptation , 1997 .

[6]  S. Osher,et al.  Algorithms Based on Hamilton-Jacobi Formulations , 1988 .

[7]  Peter H. Siegel,et al.  Fast high-resolution terahertz radar imaging at 25 meters , 2010, Defense + Commercial Sensing.

[8]  H. Quast,et al.  Towards real-time active THz range imaging for security applications , 2009, 2009 International Conference on Electromagnetics in Advanced Applications.

[9]  Yosef Pinhasi,et al.  Linear FM radar operating in the Tera-Hertz regime for concealed objects detection , 2009, 2009 IEEE International Conference on Microwaves, Communications, Antennas and Electronics Systems.

[10]  Chin-Hui Lee,et al.  A maximum-likelihood approach to stochastic matching for robust speech recognition , 1996, IEEE Trans. Speech Audio Process..

[11]  J. Sethian,et al.  Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations , 1988 .

[12]  Geoffrey J. McLachlan,et al.  Maximum likelihood clustering via normal mixture models , 1996, Signal Process. Image Commun..

[13]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[14]  T. Goyette,et al.  High Resolution Imaging using 325 GHz and 1 . 5 THz Transceivers , 2004 .

[15]  H.B. Wallace,et al.  Standoff Detection of Weapons and Contraband in the 100 GHz to 1 THz Region , 2007, IEEE Transactions on Antennas and Propagation.

[16]  Jan Flusser,et al.  Image registration methods: a survey , 2003, Image Vis. Comput..

[17]  Ralf Henneberger,et al.  Fast Active THz Cameras with Ranging Capabilities , 2009 .

[18]  Thomas E. Hall,et al.  Active millimeter-wave standoff and portal imaging techniques for personnel screening , 2009, 2009 IEEE Conference on Technologies for Homeland Security.

[19]  Patrick Lj Valdez,et al.  Standoff concealed weapon detection using a 350-GHz radar imaging system , 2010, Defense + Commercial Sensing.

[20]  R Appleby,et al.  Passive millimetre–wave imaging and how it differs from terahertz imaging , 2004, Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.