Eigenimage Filtering on Sudoeksa Field Data to Reduce Clutters
J. Hwang, D. Min, D. Kim and Y. Yoo
Event name: 24th European Meeting of Environmental and Engineering Geophysics
Session: Ground Penetrating Radar A
Publication date: 09 September 2018
Info: Extended abstract, PDF ( 516.1Kb )
Price: € 20
We compare the background-removal and eigenimage-filtering techniques in the perspective of suppressing clutters while preserving horizontal subsurface target reflections. The background-removal technique is a simple but powerful technique to remove laterally invariant clutters. Therefore, it is widely used in GPR image processing softwares. However, in case horizontal subsurface targets exist, the background-removal technique has the risk of damaging the target reflection events. One of the alternatives to this background-removal technique is the highpass-eigenimage-filtering technique. In some literatures, the effectiveness of the eigenimage-filtering technique has been proven for synthetic data sets. In this study, we compare the eigenimage-filtering technique with the background-removal technique for the field data set acquired at the testbed in Sudeoksa, Korea, for which we already have the information of subsurface target materials and locations. Through this study, we show the effectiveness of the eigenimage-filtering technique in revealing the horizontal subsurface target image.