Weighted Total Variation Joint Group Sparse Regularization for Imaging of Explosion Energy Field Attenuation on Shallow Subsurface

In the reconstruction process of subsurface explosion energy fields, several challenges arise. The complexity of shallow subsurface media significantly complicates the process. Additionally, the high explosive power of penetration-type ammunition introduces further difficulties. Limited detection sensors and their insufficient survivability worsen the situation. These factors collectively result in ill-conditioned and sparse tomographic matrices, which ultimately lead to low reconstruction accuracy. This article proposes a method of weighted total variation (WTV) joint group sparse regularization for inversion imaging of explosion energy fields under sparse measurement conditions. The WTV method is first applied to better constrain image edges by weighting image gradients and setting learnable parameters related to the data space structure. In order to preserve the detailed information of the image, we identify similarities among nonlocal subblocks of explosion energy field data based on the group sparse regularization method using low-rank constraints through block matching. Simulations and experimental validations, compared to existing algorithms, are provided to show that the proposed approach enhances the reconstruction accuracy and that the reconstruction error of this method decreases to 7%.