An Efficient Multiscale and Multidimensional Consensus for Aerial Images Nonrigid Registration

In this letter, we introduce a point set registration model based on multiscale and multidimensional consensus (MMC), designed for aerial images with varying geometric structures. In terms of local spatial partitioning, MMC utilizes four scales. The first three partitioning methods focus on grid division, partitioning the area into $10\times N$ , $20\times N$ , and $40\times N$ ; while the final approach pivots on neighborhood segmentation, with each feature point serving as the focal point, encompassing 16 adjacent points. In the realm of consensus, MMC integrates three dimensions: grid warping consensus, projected residual consensus, and neighbor deviation consensus. Our experimental results indicate that in handling repetitive textures, MMC demonstrates superior capability compared to existing state-of-the-art algorithms; in terms of processing speed and nonrigid registration capability, MMC slightly lags behind existing advanced algorithms but excels in accuracy.

I. Introduction

Point set registration serves as the foundation for various tasks, such as UAV visual navigation, change detection [1], image stitching [2], 3-D mapping [3], and image fusion [4]. With the continuous advancement of UAV technology and the gradual relaxation of flight restrictions, the utilization of UAVs in urban environments is becoming increasingly prevalent. When conducting low-altitude reconnaissance and mapping in urban areas, small UAVs face challenges due to their low shooting perspective. Complex 3-D structures, such as buildings, roads, and vegetation, exhibit multigeometric plane issues when captured from different shooting positions and angles.