SFN-YOLO: Towards Free-Range Poultry Detection via Scale-aware Fusion Networks

2025-09-21

AI Summary

To address the challenges of multi-scale poultry detection, severe occlusion, and dynamically complex backgrounds in free-range environments, this work proposes a lightweight and efficient multi-scale feature fusion framework. Methodologically, we design a scale-aware feature fusion mechanism, construct M-SCOPE—the first multi-scale poultry detection dataset tailored to free-range scenarios—and integrate a local-global contextual interaction module into the YOLO architecture to jointly optimize fine-grained detail enhancement and semantic modeling. Experimental results demonstrate that the proposed model achieves an mAP of 80.7% with only 7.2M parameters—reducing parameter count by 35.1% compared to the baseline—while exhibiting strong cross-domain generalization and real-time inference capability. This framework provides a practical, deployable solution for intelligent poultry farming.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Detecting and localizing poultry is essential for advancing smart poultry farming. Despite the progress of detection-centric methods, challenges persist in free-range settings due to multiscale targets, obstructions, and complex or dynamic backgrounds. To tackle these challenges, we introduce an innovative poultry detection approach named SFN-YOLO that utilizes scale-aware fusion. This approach combines detailed local features with broader global context to improve detection in intricate environments. Furthermore, we have developed a new expansive dataset (M-SCOPE) tailored for varied free-range conditions. Comprehensive experiments demonstrate our model achieves an mAP of 80.7% with just 7.2M parameters, which is 35.1% fewer than the benchmark, while retaining strong generalization capability across different domains. The efficient and real-time detection capabilities of SFN-YOLO support automated smart poultry farming. The code and dataset can be accessed at https://github.com/chenjessiee/SFN-YOLO.

Problem

• Detecting multiscale poultry targets in free-range farming environments
• Addressing obstructions and complex backgrounds in poultry detection
• Improving detection accuracy while reducing computational parameters significantly

Innovation

• Scale-aware fusion combining local and global features
• Lightweight architecture with 7.2M parameters
• Novel M-SCOPE dataset for free-range conditions