Table 5 Classification accuracy of the proposed YOLOv5s-CS model and YOLOv5s model for training the garbage dataset.
From: A real-time rural domestic garbage detection algorithm with an improved YOLOv5s network model
Target | YOLOv5s | YOLOv5s-CS | |||||
|---|---|---|---|---|---|---|---|
Precision | Recall | mAP@0.5 | Precision | Recall | mAP@0.5 | ||
Domestic garbage | Apple core | 0.991 | 0.998 | 0.997 | 0.99 | 0.998 | 0.997 |
Toothbrush | 0.891 | 0.779 | 0.86 | 0.985 | 0.881 | 0.950 | |
Vegetable leaf | 0.984 | 0.919 | 0.974 | 0.999 | 0.883 | 0.987 | |
Watermelon peel | 0.746 | 0.798 | 0.809 | 0.911 | 0.771 | 0.826 | |
Hazardous garbage | Battery | 0.933 | 0.854 | 0.891 | 0.94 | 0.894 | 0.847 |
Pencil | 0.875 | 0.872 | 0.875 | 0.972 | 0.92 | 0.992 | |
Button battery | 0.952 | 0.941 | 0.967 | 0.98 | 0.961 | 0.992 | |
Mobile phone | 0.955 | 0.809 | 0.944 | 0.987 | 0.977 | 0.976 | |
Recyclable garbage | Book | 0.869 | 0.792 | 0.852 | 0.923 | 0.835 | 0.886 |
Trousers | 0.966 | 0.999 | 0.995 | 0.982 | 0.973 | 0.935 | |
T-shirt | 0.969 | 0.942 | 0.969 | 0.981 | 0.982 | 0.995 | |
Other garbage | Capsule | 0.924 | 0.86 | 0.922 | 0.986 | 0.989 | 0.975 |
Remote control | 0.782 | 0.923 | 0.887 | 0.999 | 0.962 | 0.928 | |
