A Learning-based Optimization Algorithm:Image Registration Optimizer Network

TL;DR

This paper introduces IRON, a learning-based image registration optimizer that predicts the global optimal transformation parameters in a single iteration, outperforming classical methods in accuracy and efficiency.

Contribution

The paper presents a novel neural network architecture, IRON, capable of directly predicting global optima for image registration from a 3D tensor input, enabling rapid and accurate optimization.

Findings

IRON achieves higher accuracy than classical algorithms.

IRON has lower root mean square error (RMSE).

IRON demonstrates improved efficiency in image registration.

Abstract

Remote sensing image registration is valuable for image-based navigation system despite posing many challenges. As the search space of registration is usually non-convex, the optimization algorithm, which aims to search the best transformation parameters, is a challenging step. Conventional optimization algorithms can hardly reconcile the contradiction of simultaneous rapid convergence and the global optimization. In this paper, a novel learning-based optimization algorithm named Image Registration Optimizer Network (IRON) is proposed, which can predict the global optimum after single iteration. The IRON is trained by a 3D tensor (9x9x9), which consists of similar metric values. The elements of the 3D tensor correspond to the 9x9x9 neighbors of the initial parameters in the search space. Then, the tensor's label is a vector that points to the global optimal parameters from the initial parameters. Because of the special architecture, the IRON could predict the global optimum directly for any initialization. The experimental results demonstrate that the proposed algorithm performs better than other classical optimization algorithms as it has higher accuracy, lower root of mean square error (RMSE), and more efficiency. Our IRON codes are available for further study.https://www.github.com/jaxwangkd04/IRON