Rotation and scale invariant template matching applied to buried object discrimination in GPR data

In this study, a template matching approach to buried object discrimination problem is proposed over ground penetrating radar (GPR) B-scan images. The technique is scale invariant, which compensates for the change in the swinging speed of the detector. It is also rotation invariant to some extent, which reduces the number of templates to be used by compensating for the change in the scanning direction. The algorithm is tested on real GPR data and results are observed to be promising.

[1]  Joseph N. Wilson,et al.  Detecting landmines with ground-penetrating radar using feature-based rules, order statistics, and adaptive whitening , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[2]  Jiulun Fan,et al.  Minimum error thresholding: A note , 1997, Pattern Recognit. Lett..

[3]  Bülent Sankur,et al.  Survey over image thresholding techniques and quantitative performance evaluation , 2004, J. Electronic Imaging.

[4]  David J. Daniels,et al.  Surface-Penetrating Radar , 1996 .

[5]  Mark S. Drew,et al.  Full search content independent block matching based on the fast Fourier transform , 2002, Proceedings. International Conference on Image Processing.

[6]  Feng Wu,et al.  Very Fast Template Matching , 2002, ECCV.

[7]  HyunWook Park,et al.  Automatic target detection using binary template matching , 2005 .

[8]  Hai Tao,et al.  Fast Multi-scale Template Matching Using Binary Features , 2007, 2007 IEEE Workshop on Applications of Computer Vision (WACV '07).

[9]  D. Daniels.  Ground Penetrating Radar , 2005 .

[10]  Mehmet Sezgin.  Two dimensional template matching method for buried object discrimination in GPR data , 2009, Defense + Commercial Sensing.

[11]  Josef Kittler,et al.  Minimum error thresholding , 1986, Pattern Recognit..

[12]  Edward H. Adelson,et al.  PYRAMID METHODS IN IMAGE PROCESSING. , 1984 .