MS-HLMO: Multi-scale Histogram of Local Main Orientation for Remote Sensing Image Registration

TL;DR

This paper introduces MS-HLMO, a novel feature-based algorithm for remote sensing image registration that effectively handles intensity, rotation, and scale differences by using multi-scale histograms of local orientations.

Contribution

The paper proposes MS-HLMO, a new multi-scale feature descriptor and matching strategy tailored for multi-source remote sensing image registration, demonstrating superior performance.

Findings

MS-HLMO outperforms existing algorithms in effectiveness.

MS-HLMO shows strong generalization across diverse scenes.

The simplified MS-HLMO+ maintains high accuracy.

Abstract

Multi-source image registration is challenging due to intensity, rotation, and scale differences among the images. Considering the characteristics and differences of multi-source remote sensing images, a feature-based registration algorithm named Multi-scale Histogram of Local Main Orientation (MS-HLMO) is proposed. Harris corner detection is first adopted to generate feature points. The HLMO feature of each Harris feature point is extracted on a Partial Main Orientation Map (PMOM) with a Generalized Gradient Location and Orientation Histogram-like (GGLOH) feature descriptor, which provides high intensity, rotation, and scale invariance. The feature points are matched through a multi-scale matching strategy. Comprehensive experiments on 17 multi-source remote sensing scenes demonstrate that the proposed MS-HLMO and its simplified version MS-HLMO$^+$ outperform other competitive registration algorithms in terms of effectiveness and generalization.