A General Destriping Framework for Remote Sensing Images Using Flatness Constraint

TL;DR

This paper introduces a versatile destriping framework for remote sensing images that employs a novel flatness constraint to effectively characterize stripe noise, enabling the use of various regularizations within a unified convex optimization approach.

Contribution

It proposes a general destriping method with a new stripe noise characterization called flatness constraint, adaptable to different image regularizations and solved efficiently with a primal-dual splitting algorithm.

Findings

Effective destriping demonstrated on hyperspectral images and infrared videos.

Flexible framework accommodating various regularizations.

Strong stripe noise characterization via flatness constraint.

Abstract

Removing stripe noise, i.e., destriping, from remote sensing images is an essential task in terms of visual quality and subsequent processing. Most existing destriping methods are designed by combining a particular image regularization with a stripe noise characterization that cooperates with the regularization, which precludes us to examine and activate different regularizations to adapt to various target images. To resolve this, two requirements need to be considered: a general framework that can handle a variety of image regularizations in destriping, and a strong stripe noise characterization that can consistently capture the nature of stripe noise, regardless of the choice of image regularization. To this end, this paper proposes a general destriping framework using a newly-introduced stripe noise characterization, named flatness constraint, where we can handle various regularization functions in a unified manner. Specifically, we formulate the destriping problem as a nonsmooth convex optimization problem involving a general form of image regularization and the flatness constraint. The constraint mathematically models that the intensity of each stripe is constant along one direction, resulting in a strong characterization of stripe noise. For solving the optimization problem, we also develop an efficient algorithm based on a diagonally preconditioned primal-dual splitting algorithm (DP-PDS), which can automatically adjust the stepsizes. The effectiveness of our framework is demonstrated through destriping experiments, where we comprehensively compare combinations of a variety of image regularizations and stripe noise characterizations using hyperspectral images (HSI) and infrared (IR) videos.