Abstract：In view of the presence of a large amount of clutter noise mainly composed of surface direct waves in the original GPR images, In order to extract target signals effectively, an adaptive bilateral filtering method based on wavelet transform and K-means singular value decomposition is proposed. The original radar data wavelet decomposition, then K-SVD transform sparse coding and update the atoms, and get the updated sparse coefficient and small dictionary, finally will be small, the reconstructed image wavelet inverse transformation to adaptive bilateral filtering to get the image after image, to implement the wavelet transform, K-SVD algorithm and adaptive bilateral filtering method complementary advantages. The experimental results show that the proposed method has good clutter suppression effect, and the comparison with other methods shows that the proposed method has better performance in peak signal-to-noise ratio and target image entropy, and has certain application value in practical radar survey tasks.