This paper addresses the critical issue of early detection and rapid response to forest fires, which pose significant threats to ecosystems and human societies. To improve the efficiency and real-time performance of smoke detection on edge devices, the authors propose an efficient and lightweight forest smoke detection model based on YOLOv8. The key contributions of the paper include: 1. **Efficient and Lightweight Model**: The model is designed to detect and localize smoke in the early stages of forest fire spread, suitable for edge devices with limited computational resources. 2. **BiFPN Module**: A bidirectional feature pyramid network (BiFPN) is introduced to enhance feature fusion capability, improving the model's performance in detecting forest smoke. 3. **SimAM Attention Mechanism**: A simple and parameter-free attention mechanism (SimAM) is introduced to suppress redundant information and enhance the model's attention to essential features, reducing computational complexity. 4. **Focal Modulation**: Focal modulation is used to increase the focus on hard-to-detect smoke while improving the model's running speed. The experimental results show that the proposed model achieves a mean average precision (mAP) of 90.1%, a 3% improvement over the original YOLOv8s model. The number of parameters and computational complexity are reduced by 30.07% and 10.49%, respectively. The model outperforms other mainstream models in the self-built forest smoke detection dataset and has potential in practical application scenarios. The paper also discusses the impact of different attention mechanisms and detection models, and provides insights into the effectiveness of the proposed improvements.This paper addresses the critical issue of early detection and rapid response to forest fires, which pose significant threats to ecosystems and human societies. To improve the efficiency and real-time performance of smoke detection on edge devices, the authors propose an efficient and lightweight forest smoke detection model based on YOLOv8. The key contributions of the paper include: 1. **Efficient and Lightweight Model**: The model is designed to detect and localize smoke in the early stages of forest fire spread, suitable for edge devices with limited computational resources. 2. **BiFPN Module**: A bidirectional feature pyramid network (BiFPN) is introduced to enhance feature fusion capability, improving the model's performance in detecting forest smoke. 3. **SimAM Attention Mechanism**: A simple and parameter-free attention mechanism (SimAM) is introduced to suppress redundant information and enhance the model's attention to essential features, reducing computational complexity. 4. **Focal Modulation**: Focal modulation is used to increase the focus on hard-to-detect smoke while improving the model's running speed. The experimental results show that the proposed model achieves a mean average precision (mAP) of 90.1%, a 3% improvement over the original YOLOv8s model. The number of parameters and computational complexity are reduced by 30.07% and 10.49%, respectively. The model outperforms other mainstream models in the self-built forest smoke detection dataset and has potential in practical application scenarios. The paper also discusses the impact of different attention mechanisms and detection models, and provides insights into the effectiveness of the proposed improvements.