Recognizing the Damaged Surface Parts of Cars in the Real Scene Using a Deep Learning Framework

Table 3

Quantitative assessment based on recall, precision, and IoU for detecting different damaged parts of a car.

Parts	Precision (%)	Recall (%)	IoU
Windshield	93	94	0.91
Trunk lid	96	94	0.93
Roof	92	95	0.92
Front bumper	92	90	0.89
Rear bumper	91	92	0.88
Front doors	95	92	0.91
Back doors	94	90	0.91
Quarters	90	91	0.87
Fenders	93	90	0.88
Hood	92	94	0.89