In this study we propose a novel high-resolution filtering method based on non-convex regularization of the Radon Transform and demonstrate several possible applications. The Radon transfer sums signals exhibiting linear, parabolic or hyperbolic moveout in the original domain (t-Δ domain) to a single event in the new domain (Radon domain). Consequently, in the Radon domain, the target signals can be isolated easily from some types of noise. A high-resolution Radon Transform can be achieved by applying least-square inversion in the frequency domain. To further enhance the spatial resolution in the Radon domain and decrease the computational cost, a pre-processing method was developed in our group (Aharchaou and Levander 2016) based on the linear Radon transform in the frequency domain implemented with compressive sensing theory. The compressive sensing approach helps recover the sparsest solutions in the Radon domain for underdetermined inverse problems. Instead of least square minimization, a non-convex minimization (Lp regularization with 0<p<1) is used to regularize the inverse problem. In this study, we demonstrate the ability of achieving high-resolution seismic image filtering by non-convex regularization of the Radon Transform on two different setups: 1) receiver functions (RF) in receiver gathers, 2) active source shot gather data. In the first part of the study, we focus on studying the lithosphere-asthenosphere boundary (LAB). The lithosphere-asthenosphere boundary (LAB) separates the rigid, cold lithosphere characterized by a conductive thermal regime from the underlying, convecting mantle asthenosphere. The LAB is also a rheological discontinuity that marks differential motion between tectonic plates and the underlying mantle. Studying this boundary will help us to understand plate motions, tectonics, mantle convection, and properties of the upper mantle. Here, we apply this novel high-resolution filtering method on Ps RFs to remove the crustal reverberations and to enhance the LAB signals. Our results show that the quality of receiver function images for both synthetic data and broadband USArray seismic data can be improved by this approach. We further explored the possibility of filtering sedimentary basin reverberations in receiver functions so that the Mohorovičić discontinuity (Moho) and other events can be more easily identified. In the second part of the study, we apply non-convex regularized Radon Transform (NCR-RT) filtering on active source seismic data from the iMUSH (Imaging Magma Under St. Helens) project. The results demonstrate that after applying filters in Radon domain to remove S-wave and surface waves, Moho signals can be detected more precisely.