[Retracted] Efficient Object Detection Algorithm in Kitchen Appliance Scene Images Based on Deep Learning

Table 2

The results of all variants which were trained under the same conditions. The respective AP values for categories Text, Add, and Sub are used as indicators for evaluating each model.

Networks	Only 33	Only 55	Only 77	Adding 11	Our
Frame diagram	Figure 7(a)	Figure 7(b)	Figure 7(c)	Figure 7(d)	Figure 3
Text (AP)	0.9090	0.9091	0.9071	0.9081	0.9085
Add (AP)	0.9147	0.9920	0.9970	0.9923	0.9986
Sub (AP)	0.7759	0.7642	0.7394	0.7042	0.7882