A Mutual-Structure Weighted Sub-Pixel Multimodal Optical Remote Sensing Image Matching Method

TL;DR

This paper introduces a novel coarse-to-fine framework called PCWLAD for sub-pixel matching of multimodal optical remote sensing images, effectively reducing structural noise and improving accuracy across various datasets.

Contribution

The paper proposes PCWLAD, a new mutual-structure weighted method that enhances inter-modal structural consistency and achieves superior sub-pixel matching accuracy.

Findings

Outperforms eight state-of-the-art methods in accuracy

Achieves approximately 0.4 pixel average matching error

Demonstrates robustness across diverse multimodal datasets

Abstract

Sub-pixel matching of multimodal optical images is a critical step in combined application of multiple sensors. However structural noise and inconsistencies arising from variations in multimodal image responses usually limit the accuracy of matching. Phase congruency mutual-structure weighted least absolute deviation (PCWLAD) is developed as a coarse-to-fine framework. In the coarse matching stage, we preserve the complete structure and use an enhanced cross-modal similarity criterion to mitigate structural information loss by PC noise filtering. In the fine matching stage, a mutual-structure filtering and weighted least absolute deviation-based is introduced to enhance inter-modal structural consistency and accurately estimate sub-pixel displacements adaptively. Experiments on three multimodal datasets-Landsat visible-infrared, short-range visible-near-infrared, and UAV optical image pairs demonstrate that PCWLAD consistently outperforms eight state-of-the-art methods, achieving an average matching accuracy of approximately 0.4 pixels. The software and datasets are publicly available at https://github.com/huangtaocsu/PCWLAD.