Ground penetrating radar (GPR) is a non-destructive and non-intrusive subsurface investigation equipment which has been widely used in the field of civil engineering. As the GPR signals are easily corrupted by noise, so it's always difficult to detect the multi-layer reflection signal of the pavement. To address the problem,In this paper we present the wavelet transform modulus maximum de-noising and Matched Filter (MF) method for clutter reduction and layer-interface automatic detection respectively. The experimental results based on simulated data using FDTD show that under the Low SNR condition, the presented algorithm outperform the MF algorithm alone.