Edge detectors based telegraph total variational model for image filtering

Abstract

For the existing issues of edge blur and uncertainty of parameter selection during image filtering, a novel telegraph total variational PDE model based on edge detector is proposed. We propose image structure tensor as an edge detector to control smoothing process and keep more detail features. The proposed model takes advantages of both telegraph and total variational model, which is edge preserving and robust to noise. Experimental results illustrate the effectiveness of the proposed model and demonstrate that our algorithm competes favorably with state of the-art approaches in terms of producing better denoising results.