Research Article

A Robust and Lightweight Detector for Ship Target with Complex Background in SAR Image

Table 7

Comparison of performance indexes of datasets with different pruning rates (pruning model is network reconstructed model).

ModelSSDDHRSID
Precision (%)Recall (%)AP (%) (%)Precision (%)Recall (%)AP (%) (%)

0.1 prune96.795.698.896.193.888.094.190.8
0.2 prune96.595.298.795.893.287.694.090.3
0.3 prune96.694.998.695.792.687.893.890.1
0.4 prune96.395.298.195.792.987.293.390.0
0.5 prune96.194.097.595.091.785.992.788.7
0.6 prune95.192.997.194.091.185.192.188.0
0.7 prune91.995.596.993.791.283.590.787.2
0.8 prune94.789.096.791.888.979.388.083.8
0.9 prune92.882.492.487.386.072.982.178.9