Figure 1.
YOLOv7 network architecture diagram
Figure 2.
Schematic diagram of Multi_Concat_Block module
Figure 3.
Schematic diagram of Transition Block module
Figure 4.
Sketch of SPPCSPC structure
Figure 5.
Feature layer shape change map
Figure 6.
Map of the location of the introduction of the attention mechanism
Figure 7.
Roadmap for system realization
TABLE I.
Table of fruit types and corresponding number of pictures
| Name and number of vegetables | Name and number of fruit |
|---|---|
| Cabbage (200) | Apple (200) |
| Capsicum (200) | Banana (200) |
| Carrot (200) | Pear (200) |
| Cauliflower (200) | Pineapple (200) |
| Corn (200) | Pomegranate (200) |
| Eggplant (200) | Grapes (200) |
| Cabbage (200) | Apple (200) |
Figure 8.
Confusion matrix diagram
Figure 9.
F1 score graph
Figure 10.
P_curve
Figure 11.
PR_curve
Figure 12.
R_curve
Figure 13.
Apple experiment result
Figure 14.
Results of Onion and Carrot experiments
TABLE II.
Comparison table of detection accuracy
| Type | Evaluation metrics | ||
|---|---|---|---|
| Detection Times | mAP/% (Pre-improved) | mAP/% (Improved) | |
| Apple | 30 | 0.79 | 0.85 |
| Banana | 30 | 0.74 | 0.77 |
| Pear | 30 | 0.79 | 0.81 |
| Pineapple | 30 | 0.81 | 0.79 |
| Pomegranate | 30 | 0.68 | 0.68 |
| Grapes | 30 | 0.50 | 0.57 |
| Watermelon | 30 | 0.73 | 0.78 |
| Cabbage | 30 | 0.89 | 0.91 |
| Capsicum | 30 | 0.55 | 0.58 |
| Carrot | 30 | 0.83 | 0.89 |
| Cauliflower | 30 | 0.55 | 0.64 |
| Corn | 30 | 0.46 | 0.45 |
| Eggplant | 30 | 0.74 | 0.71 |
| Onion | 30 | 0.88 | 0.95 |