Multi-temporal Adaptive Red-Green-Blue and Long-Wave Infrared Fusion for You Only Look Once-Based Landmine Detection from Unmanned Aerial Systems

2025-12-23

AI Summary

Addressing the challenge of weak and environmentally sensitive thermal signatures of surface-laid landmines in humanitarian demining, this paper proposes a multi-temporal adaptive RGB–long-wave infrared (LWIR) fusion detection method deployed on unmanned aerial systems (UAS). The approach introduces a novel dynamic-weighted feature fusion mechanism, integrated with multi-temporal thermal data augmentation and a UAS-coordinated perception framework. Systematic analysis identifies 10–30% LWIR channel contribution and 5–10 m flight altitude as optimal operational parameters; cross-seasonal training is empirically shown to enhance thermal robustness, improving mean average precision (mAP) by 1.8–9.6%. Implemented on YOLOv11, the method achieves 86.8% mAP under optimal configuration, with detection recall (AT) and precision (AP) of 61.9% and 19.2%, respectively, and trains 17.7× faster than RF-DETR. This work establishes a scalable, thermally robust paradigm for real-time landmine detection in complex field environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Landmines remain a persistent humanitarian threat, with 110 million actively deployed mines across 60 countries, claiming 26,000 casualties annually. This research evaluates adaptive Red-Green-Blue (RGB) and Long-Wave Infrared (LWIR) fusion for Unmanned Aerial Systems (UAS)-based detection of surface-laid landmines, leveraging the thermal contrast between the ordnance and the surrounding soil to enhance feature extraction. Using You Only Look Once (YOLO) architectures (v8, v10, v11) across 114 test images, generating 35,640 model-condition evaluations, YOLOv11 achieved optimal performance (86.8% mAP), with 10 to 30% thermal fusion at 5 to 10m altitude identified as the optimal detection parameters. A complementary architectural comparison revealed that while RF-DETR achieved the highest accuracy (69.2% mAP), followed by Faster R-CNN (67.6%), YOLOv11 (64.2%), and RetinaNet (50.2%), YOLOv11 trained 17.7 times faster than the transformer-based RF-DETR (41 minutes versus 12 hours), presenting a critical accuracy-efficiency tradeoff for operational deployment. Aggregated multi-temporal training datasets outperformed season-specific approaches by 1.8 to 9.6%, suggesting that models benefit from exposure to diverse thermal conditions. Anti-Tank (AT) mines achieved 61.9% detection accuracy, compared with 19.2% for Anti-Personnel (AP) mines, reflecting both the size differential and thermal-mass differences between these ordnance classes. As this research examined surface-laid mines where thermal contrast is maximized, future research should quantify thermal contrast effects for mines buried at varying depths across heterogeneous soil types.

Problem

• Detects surface-laid landmines using UAS with RGB and thermal fusion
• Optimizes YOLO models for accuracy and speed in mine detection
• Evaluates detection performance across different mine types and conditions

Innovation

• Multi-temporal adaptive RGB and LWIR fusion for UAS landmine detection
• YOLOv11 achieves optimal performance with thermal fusion parameters
• Aggregated multi-temporal training outperforms season-specific approaches