Polished Surface Defect Detection Based on Intelligent Surface Analysis

Fig. 11. Comparison of the results between the proposed detection model and other mainstream detection models on the UPP-DET dataset. (a) Overall comparison; (b) partial comparison

Download full size

Fig. 12. Test results of defect detection model

Download full size

Table 1. Basic parameters of the experiment
View table
Table 1. Basic parameters of the experiment
Parameter UPP-CLS UPP-DET
Batch size 8 4
Epoch 50 15
Initial learning rate 0.001 0.0002
Weight decay ‒ 0.05
Optimizer Adam AdamW
Regularization L2 weight decay ‒

Table 2. Module performance analysis in the FIFE Block
View table
Table 2. Module performance analysis in the FIFE Block
Vector ordering SC SE Accuracy /%
L C
‒ ‒ ‒ ‒ 75.18
√ ‒ ‒ ‒ 74.39
‒ √ ‒ ‒ 78.07
‒ √ √ ‒ 78.24
‒ √ √ √ 79.47

Table 3. Comparative analysis of each model embedded in FIFE block

View table

Table 3. Comparative analysis of each model embedded in FIFE block

Image branch backbone	FIFE	Accuracy /%	Sensitivity /%	FNR /%	TNR /%	Specificity /%
InceptionV3^［23］	‒	70.47	68.83	31.17	73.17	26.83
InceptionV3^［23］	√	74.44	71.23	28.77	77.70	22.30
ResNet-50^［21］	‒	75.18	73.64	26.36	77.88	22.12
ResNet-50^［21］	√	79.47	76.12	23.88	76.27	23.73
ResNet-101^［21］	‒	78.63	76.03	23.97	80.00	20.00
ResNet-101^［21］	√	83.98	79.68	20.32	83.21	16.79
MobileNetV3^［24］	‒	70.09	72.11	27.89	71.67	28.33
MobileNetV3^［24］	√	75.69	72.99	27.01	76.35	23.65
ResNext-50^［25］	‒	78.87	77.69	22.31	80.04	19.96
ResNext-50^［25］	√	82.21	77.15	22.85	84.76	15.24
EfficientNet-b2^［19］	‒	80.41	78.25	21.75	82.67	17.33
EfficientNet-b2^［19］	√	84.86	80.89	19.11	86.38	13.62
EfficientNet-b4^［19］	‒	82.11	79.97	20.03	84.64	15.36
EfficientNet-b4^［19］	√	85.36	81.49	18.51	87.72	12.28

Table 4. Performance analysis of each model embedded in FIFE block
View table
Table 4. Performance analysis of each model embedded in FIFE block
AD Mixup DC mAP
— — — 0.697
√ — — 0.721
√ √ — 0.729
√ √ √ 0.740

Table 5. Comparison results between proposed model and other detection models

View table

Table 5. Comparison results between proposed model and other detection models

Stage	Method	Backbone	MST	Scratch	Pit	mAP	FPS /（frame·s^-1）
One-stage	SSD300^［27］	VGG-16	‒	0.363	0.335	0.349	‒
	SSD512^［27］	VGG-16	‒	0.581	0.593	0.587	‒
	RetinaNet^［28］	ResNet-50	‒	0.775	0.749	0.762	‒
	Yolov3^［29］	DarkNet-53	‒	0.632	0.594	0.613	‒
	Yolov5	DarkNet-53	‒	0.816	0.782	0.799	‒
Two-stage	Faster R-CNN^［30］	ResNet-50	√	0.729	0.751	0.740	28.3
	Dynamic R-CNN^［31］	ResNet-50	√	0.816	0.798	0.807	24.9
	Cascade R-CNN^［18］	ResNet-50	√	0.815	0.822	0.819	22.7
	Proposed	EfficientNet-b4	√	0.867	0.840	0.854	21.1

Table 6. IoU threshold analysis results at each stage of defect detection model detection head

View table

Table 6. IoU threshold analysis results at each stage of defect detection model detection head

No.	$T_{I o U}$	mAP
Baseline	［0.5，0.6，0.7］	0.854
1	［0.2，0.6，0.7］	0.841
2	［0.3，0.6，0.7］	0.853
3	［0.4，0.6，0.7］	0.856
4	［0.4，0.5，0.6］	0.851
5	［0.4，0.5，0.7］	0.862
6	［0.4，0.5，0.8］	0.839

Tools

Get Citation

Copy Citation Text

Zihao Li, Fengzhou Fang, Zhonghe Ren, Gaofeng Hou. Polished Surface Defect Detection Based on Intelligent Surface Analysis[J]. Laser & Optoelectronics Progress, 2023, 60(24): 2412006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites