Correspondence A Total Variation-Based Algorithm for Pixel-Level Image Fusion

Abstract—In this paper, a total variation (TV) based approach is pro-posed for pixel-level fusion to fuse images acquired using multiple sensors. In this approach, fusion is posed as an inverse problem and a locally affine model is used as the forward model. A TV seminorm based approach in conjunction with principal component analysis is used iteratively to esti-mate the fused image. The feasibility of the proposed algorithm is demon-strated on images from computed tomography (CT) and magnetic reso-nance imaging (MRI) as well as visible-band and infrared sensors. The re-sults clearly indicate the feasibility of the proposed approach. Index Terms—Eigenvector, forward model, image fusion, inverse problem, pixel-level fusion, total variation (TV). I