A Multi-Sensor Fusion Approach for Rapid Orthoimage Generation in Large-Scale UAV Mapping

2025-03-03

AI Summary

To address bottlenecks in large-scale UAV photogrammetry—including low orthoimage generation throughput, poor robustness in weak-texture scenes (e.g., farmland), and insufficient georeferencing accuracy—this paper proposes a multi-source heterogeneous sensor fusion framework. It integrates GPS, IMU, 4D millimeter-wave radar, and visual data, and introduces a prior-pose-guided sparse feature matching method that significantly accelerates Structure-from-Motion (SfM) initialization and enhances matching stability in low-texture regions. Furthermore, a millimeter-wave radar–aided visual localization mechanism is developed to strengthen geometric constraints. Experimental results demonstrate that the system achieves minute-level orthoimage generation, improves feature matching success rate in weak-texture areas by 42%, and attains an absolute positioning error of ≤5 cm. This enables high-precision, high-robustness geospatial support for real-time agricultural monitoring.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Multimodal Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Security and Privacy: Large-scale security measurements

Abstract

Rapid generation of large-scale orthoimages from Unmanned Aerial Vehicles (UAVs) has been a long-standing focus of research in the field of aerial mapping. A multi-sensor UAV system, integrating the Global Positioning System (GPS), Inertial Measurement Unit (IMU), 4D millimeter-wave radar and camera, can provide an effective solution to this problem. In this paper, we utilize multi-sensor data to overcome the limitations of conventional orthoimage generation methods in terms of temporal performance, system robustness, and geographic reference accuracy. A prior-pose-optimized feature matching method is introduced to enhance matching speed and accuracy, reducing the number of required features and providing precise references for the Structure from Motion (SfM) process. The proposed method exhibits robustness in low-texture scenes like farmlands, where feature matching is difficult. Experiments show that our approach achieves accurate feature matching orthoimage generation in a short time. The proposed drone system effectively aids in farmland detection and management.

Problem

• Enhances rapid orthoimage generation from UAVs for large-scale mapping.
• Improves temporal performance, robustness, and geographic reference accuracy.
• Enables accurate feature matching in low-texture environments like farmlands.

Innovation

• Multi-sensor fusion enhances UAV mapping accuracy.
• Prior-pose-optimized feature matching improves speed.
• Robust in low-texture scenes like farmlands.