Detecting Wildfire Flame and Smoke through Edge Computing using Transfer Learning Enhanced Deep Learning Models

2025-01-15

AI Summary

This study addresses the challenge of real-time, low-power wildfire detection—specifically flame and smoke—on resource-constrained UAV edge devices under few-shot learning conditions. We propose a two-stage cascaded transfer learning strategy: first fine-tuning lightweight YOLOv5n and YOLOv11n models pretrained on FASDD/COCO, then deploying them on NVIDIA Jetson platforms. To our knowledge, this is the first work to systematically quantify how transfer learning impacts edge inference latency, power consumption, and energy efficiency. Experiments show that YOLOv5n achieves ~2× higher inference speed than YOLOv11n without hardware acceleration, attaining 79.2% mAP@0.5 while significantly reducing training time and improving generalization. Key contributions are: (1) an energy-efficient, few-shot wildfire detection framework tailored for edge deployment; (2) empirical analysis of transfer learning’s impact on critical edge metrics; and (3) validation of YOLOv5n’s superiority in highly resource-constrained scenarios.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Autonomous unmanned aerial vehicles (UAVs) integrated with edge computing capabilities empower real-time data processing directly on the device, dramatically reducing latency in critical scenarios such as wildfire detection. This study underscores Transfer Learning's (TL) significance in boosting the performance of object detectors for identifying wildfire smoke and flames, especially when trained on limited datasets, and investigates the impact TL has on edge computing metrics. With the latter focusing how TL-enhanced You Only Look Once (YOLO) models perform in terms of inference time, power usage, and energy consumption when using edge computing devices. This study utilizes the Aerial Fire and Smoke Essential (AFSE) dataset as the target, with the Flame and Smoke Detection Dataset (FASDD) and the Microsoft Common Objects in Context (COCO) dataset serving as source datasets. We explore a two-stage cascaded TL method, utilizing D-Fire or FASDD as initial stage target datasets and AFSE as the subsequent stage. Through fine-tuning, TL significantly enhances detection precision, achieving up to 79.2% mean Average Precision (mAP@0.5), reduces training time, and increases model generalizability across the AFSE dataset. However, cascaded TL yielded no notable improvements and TL alone did not benefit the edge computing metrics evaluated. Lastly, this work found that YOLOv5n remains a powerful model when lacking hardware acceleration, finding that YOLOv5n can process images nearly twice as fast as its newer counterpart, YOLO11n. Overall, the results affirm TL's role in augmenting the accuracy of object detectors while also illustrating that additional enhancements are needed to improve edge computing performance.

Problem

• Wildfire Detection
• Autonomous Drones
• Energy Efficiency

Innovation

• Transfer Learning
• YOLOv5n
• Wildfire Detection