Buried Object Detection Based on Multivariate Latent Methods

Ghamry, K. A.; Hussein, W. M.; Hegazy, M.; Mahgoub, H. M.

doi:10.21608/asat.2011.23277

Buried Object Detection Based on Multivariate Latent Methods

Document Type : Original Article

Authors

Egyptian Armed Forces, Cairo, Egypt.

10.21608/asat.2011.23277

Abstract

In this paper, wideband ground penetrating radar (GPR) system and a proposed frequency domain data analysis technique are presented for the detection of shallow buried objects such as anti-personal (AP) land mines. The GPR system uses one antenna for transmitting and for receiving and. it operates from 9 GHz to11 GHz. This system is able to acquire, save and analyze the data in the frequency domain. Experiments were conducted on metallic and plastic AP mines under different depths. Principal Component Analysis (PCA) has been used for processing the data. This technique is effective for the clustering of different groups. Results show that the proposed technique can successfully differentiate between metallic and plastic mines-like targets buried in dry soil.

Keywords