SFFA-YOLO: A Multi-Scale Fusion Algorithm for Fire Smoke Detection
Open Access
- 5 February 2026
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 16 (3) , 1599
Abstract
The rapid spread of fires underscores the urgency of high-accuracy fire smoke detection for public safety, but early fires pose major challenges—small flame/smoke targets, blurred boundaries, low contrast, and complex background interference—limiting the performance of existing models. To address these issues, this paper proposes SFFA-YOLO, an engineering-oriented improved algorithm based on the YOLOv11 framework for fire smoke detection, which achieves a balanced trade-off between detection precision, real-time performance, and lightweight deployment. The model integrates three synergistic optimization modules for targeted scene adaptation: (1) the FMFA module for cross-scale feature fusion to enhance thin smoke and small flame recognition; (2) the SGCA module for joint channel-spatial feature focusing to improve target localization accuracy; (3) the SDA-Loss function for dynamic weight adjustment based on target size and clarity to stabilize small target detection. Validated on the self-built FS-Blend dataset (supplemented with difficult samples such as distant thin smoke and backlit flames), SFFA-YOLO outperforms mainstream models (YOLOv8, YOLOv9, Faster R-CNN) in key metrics. Compared with the YOLOv11s baseline, it achieves a 2.5% Precision improvement and 3.9% mAP@0.5 improvement while reducing parameters by 12.8%, confirming its reliability as a real-time fire smoke detection solution.Keywords
Funding Information
- Heilongjiang Provincial Postdoctoral Research Start-Up Fund Project (LBHQ19028)
This publication has 38 references indexed in Scilit:
- DSS-YOLO: an improved lightweight real-time fire detection model based on YOLOv8Scientific Reports, 2025
- Multiscale wildfire and smoke detection in complex drone forest environments based on YOLOv8Scientific Reports, 2025
- YOLO-EPF: Multi-scale smoke detection with enhanced pool former and multiple receptive fieldsDigital Signal Processing, 2024
- Powerful-IoU: More straightforward and faster bounding box regression loss with a nonmonotonic focusing mechanismNeural Networks, 2023
- A deep learning model using geostationary satellite data for forest fire detection with reduced detection latencyGIScience & Remote Sensing, 2022
- Fire detection in video surveillances using convolutional neural networks and wavelet transformEngineering Applications of Artificial Intelligence, 2022
- Early indoor occluded fire detection based on firelight reflection characteristicsFire Safety Journal, 2022
- Fire Detection Based on Fractal Analysis and Spatio-Temporal FeaturesFire Technology, 2021
- Convolutional neural networks based on multi-scale additive merging layers for visual smoke recognitionMachine Vision and Applications, 2018
- Fire detection using statistical color model in video sequencesJournal of Visual Communication and Image Representation, 2007