PSTAIC regularization for 2D spatiotemporal image reconstruction

TL;DR

This paper introduces an enhanced regularization method for 2D spatiotemporal image reconstruction that adaptively estimates weights within the optimization process, improving restoration quality.

Contribution

It extends the STAIC model by integrating weight estimation into the optimization, enabling more effective regularization for image restoration.

Findings

Improved restoration quality compared to existing models.

Effective adaptive weight estimation scheme.

Demonstrated robustness across different restoration scenarios.

Abstract

We propose a model for restoration of spatio-temporal TIRF images based on infimal decomposition regularization model named STAIC proposed earlier. We propose to strengthen the STAIC algorithm by enabling it to estimate the relative weights in the regularization term by incorporating it as part of the optimization problem. We also design an iterative scheme which alternatively minimizes the weight and image sub-problems. We demonstrate the restoration quality of this regularization scheme against other restoration models enabled by similar weight estimation schemes.