Pest Rat Detection and Capture System in Barn

With the development of agricultural automation and intelligence, pest rat monitoring in barns becomes more and more important. The traditional rat detection method has problems such as low precision and poor efficiency, which is difficult to meet the practical needs. In this paper, an improved method for rat detection in barn is proposed based on YOLOv5 algorithm. First, in order to improve the detection accuracy, a multi-scale feature fusion strategy is adopted, which improves the network structure in order to better capture the features of rats at different scales; second, data enhancement techniques, including image rotation, scaling, and color transformations, are used to enhance the robustness and generalization ability of the model. Finally, in order to improve the real-time detection capability, lightweight network design and quantization techniques are used to reduce the computational and parameter numbers of the model, thus achieving faster detection. The experimental results show that the improved YOLOv5 model has a greater improvement in both accuracy and detection speed compared to the original YOLOv5 model on the rat detection task. The method proposed in this paper has greater potential for application in agricultural surveillance and provides technical support for automated pest rat control.