I. Introduction
GPR is an instrument that is used for subsurface mapping using high frequency Electromagnetic (EM) waves. An attempt has been made to use GPR to detect and locate buried objects. Using GPR images we can identify metallic and non metallic buried utilities like primary mines, pipes, cables etc. Further, the GPR images can be used to obtain more quantitative information about the location, shape and size of buried objects reflecting the radar signals [1]. The images obtained using GPR are full of information, however these images contain undesired elements like: ground echo, noise and clutters [2]. Raw images obtained using GPR are called -scan. The patterns appearing in the -scan have shapes, which is obtained by the propagation of short pulses into a medium with certain electrical properties. The task of identifying buried objects using GPR can be realized in two ways. Firstly, by processing the received raw images from GPR, secondly, by improving and customizing the hardware used in GPR system to suit the needs of an application. The first option is feasible using various image processing techniques, whereas, second option is economically infeasible. The important task always is to visualize the image conveniently. Hence, to identify and automatically detect the object becomes challenging.