Improving Image Restoration with Soft-Rounding

TL;DR

This paper introduces a soft-rounding regularizer that leverages known pixel value sets to improve image restoration, especially for images with discrete pixel values, leading to better visual quality and quantitative metrics.

Contribution

The paper proposes a novel regularizer that incorporates known pixel value sets into the restoration process, extending the rounding operation to improve restoration of discrete-valued images.

Findings

Significant improvement in PSNR and SSIM for binary and pattern images.

Effective enhancement of natural image restoration.

Efficient implementation of the soft-rounding step.

Abstract

Several important classes of images such as text, barcode and pattern images have the property that pixels can only take a distinct subset of values. This knowledge can benefit the restoration of such images, but it has not been widely considered in current restoration methods. In this work, we describe an effective and efficient approach to incorporate the knowledge of distinct pixel values of the pristine images into the general regularized least squares restoration framework. We introduce a new regularizer that attains zero at the designated pixel values and becomes a quadratic penalty function in the intervals between them. When incorporated into the regularized least squares restoration framework, this regularizer leads to a simple and efficient step that resembles and extends the rounding operation, which we term as soft-rounding. We apply the soft-rounding enhanced solution to the restoration of binary text/barcode images and pattern images with multiple distinct pixel values. Experimental results show that soft-rounding enhanced restoration methods achieve significant improvement in both visual quality and quantitative measures (PSNR and SSIM). Furthermore, we show that this regularizer can also benefit the restoration of general natural images.