Abstract: A method for subsurface metallic landmine identification in air- or vehicle-borne Ground Penetrating Synthetic Aperture Radar （GPSAR） using finite difference time domain （FDTD） method combined with genetic algorithm is proposed. To identify the real object, the framework information on landmine is obtained according to the matching algorithm based on the number and relative location of one dimensional scattering center. We build the computational model for subsurface landmine using finite difference time domain method. To testify the proposed computational model, the electromagnetic characteristic on M6A1 metallic landmine buried beneath a Gaussian rough interface is analyzed. Numerical results fit real data based on actual measurement. By identifying two kinds of metallic landmines buried beneath a Gaussian rough interface, the effectiveness of the proposed method is testified.