# FF-Mamba-YOLO: An SSM-Based Benchmark for Forest Fire Detection in UAV Remote Sensing Images

## TL;DR

This paper introduces FF-Mamba-YOLO, a new framework for detecting forest fires in UAV images, which outperforms existing methods by capturing global dependencies and improving feature processing.

## Contribution

The novel FF-Mamba-YOLO framework introduces MFEBlock, MFFBlock, CFEBlock, and MGBlock to enhance global and local feature processing in forest fire detection.

## Key findings

- FF-Mamba-YOLO achieves 67.4% mAP@50 on a self-constructed forest fire image dataset.
- The model outperforms previous state-of-the-art methods in precision and mAP metrics.
- Dynamic modules like MGBlock and DySample improve adaptability and image resolution without high computational costs.

## Abstract

Timely and accurate detection of forest fires through unmanned aerial vehicle (UAV) remote sensing target detection technology is of paramount importance. However, multiscale targets and complex environmental interference in UAV remote sensing images pose significant challenges during detection tasks. To address these obstacles, this paper presents FF-Mamba-YOLO, a novel framework based on the principles of Mamba and YOLO (You Only Look Once) that leverages innovative modules and architectures to overcome these limitations. Specifically, we introduce MFEBlock and MFFBlock based on state space models (SSMs) in the backbone and neck parts of the network, respectively, enabling the model to effectively capture global dependencies. Second, we construct CFEBlock, a module that performs feature enhancement before SSM processing, improving local feature processing capabilities. Furthermore, we propose MGBlock, which adopts a dynamic gating mechanism, enhancing the model’s adaptive processing capabilities and robustness. Finally, we enhance the structure of Path Aggregation Feature Pyramid Network (PAFPN) to improve feature fusion quality and introduce DySample to enhance image resolution without significantly increasing computational costs. Experimental results on our self-constructed forest fire image dataset demonstrate that the model achieves 67.4% mAP@50, 36.3% mAP@50:95, and 64.8% precision, outperforming previous state-of-the-art methods. These results highlight the potential of FF-Mamba-YOLO in forest fire monitoring.