Ground Penetrating Radar (GPR) is a non-destructive technique used for the location of objects or interfaces buried beneath the earth's surface or located within a visually opaque structure. This research work proposes techniques for buried object discrimination for the images generated by GPR by using GPR frequency-domain spectral features. The motivation for this approach comes from the fact that landmine targets and clutter objects often have different shapes and/or composition, yielding different Energy Density Spectrum (EDS) that may be exploited for their discrimination. The proposed system comprises cascade of two stages: Pre - Processing Stage, followed by Discrimination Stage. Pre Processing eliminates the ground bounce and clutter to get an enhanced image. Enhancement is followed by Landmine or clutter Discrimination using Energy Density Spectrum. All implementations are done using MATLAB.