Research on the Application of YOLOv5 Model in Household Waste Classification and Detection

With the increasing awareness of environmental protection, intelligent household waste classification has become an important part of urban management. However, the traditional household waste classification is inefficient and prone to make mistakes. Therefore, this paper proposes a smart waste classification and detection method based on YOLOv5 to improve the accuracy and efficiency of domestic waste classification. First of all, facing the problem of unbalanced and inaccurate junk image data, the comprehensiveness of model training is ensured through data annotation and enhancement technology. Subsequently, after an in-depth comparison of multiple models, YOLOv5 was selected as the best model. Then, the model profile is adjusted for hyperparameter processing to improve the performance and generalization ability of the model ${ }^{[11}$. Fully trained and rigorously evaluated, the model performs well in the detection and classification tasks. Finally, the optimized model is applied to the actual scene to realize the detection and classification of intelligent household waste.