SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD

Liu, Juan; Feng, Jiaxin; Wang, Shujie; Ding, Yian; Guo, Jianghua; Li, Yuhang; Xue, Wenxuan; Hu, Jie

doi:10.3390/f17010093

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD

by

Juan Liu

^*,

Jiaxin Feng

,

Shujie Wang

,

Yian Ding

,

Jianghua Guo

,

Yuhang Li

,

Wenxuan Xue

and

Jie Hu

Shanxi Agricultural University, Taigu 030800, China

^*

Author to whom correspondence should be addressed.

Forests 2026, 17(1), 93; https://doi.org/10.3390/f17010093 (registering DOI)

Submission received: 25 November 2025 / Revised: 27 December 2025 / Accepted: 6 January 2026 / Published: 10 January 2026

(This article belongs to the Section Natural Hazards and Risk Management)

Download Versions Notes

Abstract

Forest ecosystems, as vital natural resources, are increasingly endangered by wildfires. Effective forest fire management relies on the accurate and early detection of small–scale flames and smoke. However, the complex and dynamic forest environment, along with the small size and irregular shape of early fire indicators, poses significant challenges to reliable early warning systems. To address these issues, this paper introduces SER–YOLOv8, an enhanced detection model based on the YOLOv8 architecture. The model incorporates the RepNCSPELAN4 module and an SPPELAN structure to strengthen multi-scale feature representation. Furthermore, to improve small target localization, the Normalized Wasserstein Distance (NWD) loss is adopted, providing a more robust similarity measure than traditional IoU–based losses. The newly designed SERDet module deeply integrates a multi–scale feature extraction mechanism with a multi-path fused attention mechanism, significantly enhancing the recognition capability for flame targets under complex backgrounds. Depthwise separable convolution (DWConv) is utilized to reduce parameters and boost inference efficiency. Experiments on the M4SFWD dataset show that the proposed method improves mAP50 by 1.2% for flames and 2.4% for smoke, with a 1.5% overall gain in mAP50–95 over the baseline YOLOv8, outperforming existing mainstream models and offering a reliable solution for forest fire prevention.

Keywords: forest fire; YOLOv8; RepNCSPELAN4; SPPELAN; NWD; SERDet

Share and Cite

MDPI and ACS Style

Liu, J.; Feng, J.; Wang, S.; Ding, Y.; Guo, J.; Li, Y.; Xue, W.; Hu, J. SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD. Forests 2026, 17, 93. https://doi.org/10.3390/f17010093

AMA Style

Liu J, Feng J, Wang S, Ding Y, Guo J, Li Y, Xue W, Hu J. SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD. Forests. 2026; 17(1):93. https://doi.org/10.3390/f17010093

Chicago/Turabian Style

Liu, Juan, Jiaxin Feng, Shujie Wang, Yian Ding, Jianghua Guo, Yuhang Li, Wenxuan Xue, and Jie Hu. 2026. "SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD" Forests 17, no. 1: 93. https://doi.org/10.3390/f17010093

APA Style

Liu, J., Feng, J., Wang, S., Ding, Y., Guo, J., Li, Y., Xue, W., & Hu, J. (2026). SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD. Forests, 17(1), 93. https://doi.org/10.3390/f17010093

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

SER-YOLOv8: An Early Forest Fire Detection Model Integrating Multi-Path Attention and NWD

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI