# A Ground-Based Visual System for UAV Detection and Altitude Measurement Deployment and Evaluation of Ghost-YOLOv11n on Edge Devices

## TL;DR

The paper introduces Ghost-YOLOv11n, a lightweight drone detection system that works efficiently on low-power devices and accurately measures drone altitude up to 30 meters.

## Contribution

Ghost-YOLOv11n reduces computational cost by 12.7% while maintaining high accuracy and enabling real-time drone detection and altitude measurement on edge devices.

## Key findings

- Ghost-YOLOv11n achieves 98.8% mAP0.5 on a dataset of 8795 images with 12.7% lower computational cost.
- The system runs at 20 FPS on a LuBanCat4 edge device with NPU acceleration.
- Altitude measurement errors remain within 10% up to 30 meters using monocular vision and EKF.

## Abstract

What are the main findings?
We propose Ghost-YOLOv11n, a lightweight UAV detector that reduces computational cost by 12.7% while achieving 98.8% mAP0.5 on a comprehensive dataset of 8795 images.The system, deployed on a low-power LuBanCat4 edge device with NPU acceleration, achieves 20 FPS and maintains altitude measurement errors within 10% up to 30 m using a monocular vision- and EKF-based approach.

We propose Ghost-YOLOv11n, a lightweight UAV detector that reduces computational cost by 12.7% while achieving 98.8% mAP0.5 on a comprehensive dataset of 8795 images.

The system, deployed on a low-power LuBanCat4 edge device with NPU acceleration, achieves 20 FPS and maintains altitude measurement errors within 10% up to 30 m using a monocular vision- and EKF-based approach.

What are the implications of the main findings?
This work provides a cost-effective, edge-deployable solution for ground-based UAV surveillance, enabling long-term, low-power operation suitable for protecting critical infrastructure.It establishes a practical benchmark for integrated detection and altitude measurement systems in real-world ground-to-air scenarios, bridging the gap between algorithm design and system-level deployment.

This work provides a cost-effective, edge-deployable solution for ground-based UAV surveillance, enabling long-term, low-power operation suitable for protecting critical infrastructure.

It establishes a practical benchmark for integrated detection and altitude measurement systems in real-world ground-to-air scenarios, bridging the gap between algorithm design and system-level deployment.

The growing threat of unauthorized drones to ground-based critical infrastructure necessitates efficient ground-to-air surveillance systems. This paper proposes a lightweight framework for UAV detection and altitude measurement from a fixed ground perspective. We introduce Ghost-YOLOv11n, an optimized detector that integrates GhostConv modules into YOLOv11n, reducing computational complexity by 12.7% while achieving 98.8% mAP0.5 on a comprehensive dataset of 8795 images. Deployed on a LuBanCat4 edge device with Rockchip RK3588S NPU acceleration, the model achieves 20 FPS. For stable altitude estimation, we employ an Extended Kalman Filter to refine measurements from a monocular ranging method based on similar-triangle geometry. Experimental results under ground monitoring scenarios show height measurement errors remain within 10% up to 30 m. This work provides a cost-effective, edge-deployable solution specifically for ground-based anti-drone applications.