Researching | Milling Surface Roughness Measurement Under Few-Shot Problem

Laser & Optoelectronics Progress, Volume. 59, Issue 23, 2324001(2022)

Milling Surface Roughness Measurement Under Few-Shot Problem

Huaian Yi^1、*, Runji Fang¹, Aihua Shu², and Enhui Lu³

Author Affiliations

¹School of Mechanical and Control Engineering, Guilin University of Technology, Guilin 541006, Guangxi, China

²School of Foreign Languages, Guilin University of Technology, Guilin 541006, Guangxi, China

³School of Mechanical Engineering, Yangzhou University, Yangzhou 225009, Jiangsu, China

show less

Abstract Get PDF(in Chinese)

Figures&Tables (15)

References (16)

Paper Information

Figures & Tables(15)

Fig. 1. GNN network structure diagram

Download full size

Fig. 2. Partial effect of data enhancement. (a) Original image; (b) adjusting contrast; (c) adjusting saturation; (d) adjusting hue; (e) translation

Download full size

Fig. 3. Experimental design flow

Download full size

Fig. 4. Machining machine and tool

Download full size

Fig. 5. Experimental setup

Download full size

Fig. 6. Texture orientation. (a) Left; (b) right

Download full size

Fig. 7. Light and dark uneven distribution and reflection phenomenon

Download full size

Fig. 8. Image pre-processing. (a) Clipping; (b) compression

Download full size

Fig. 9. Experimental results. Loss function curves and accuracy curves of (a) (b) MAML and (c) (d) GNN

Download full size

Fig. 10. Experimental results. Loss function curves and accuracy curves for cross-domain detection tasks (a) (b) A to B and (c) (d) B to A

Download full size

Table 1. Material and processing parameters

Table 1. Material and processing parameters

Material	Size /（mm×mm）	Roughness range /µm	Roughness measuring instrument	Numerical control machine
45#steel	60×40	0.6-5.0	Mitutoyo SJ-301	XHS7145
Milling cutter	Milling cutter blade	Cutting depth /mm	Feed rate /（mm·min^-1）	Spindle speed /（r·min^-1）
TAP400R100-32-6T	APMT1604PDER TR330	0.05-0.20	100-2200	700

Table 3. Sample classification and number statistics after data pre-processing
View table
Table 3. Sample classification and number statistics after data pre-processing
Set Number of samples Data enhancement Number of samples
Training set 984 Yes 9840
Validation set 288 No
Test set 432 No

Table 4. Cross-domain detection tasks
View table
Table 4. Cross-domain detection tasks
Name Training set Validation set Test set
A to B A B B
B to A B A A

Table 5. Test accuracy of MAML and GNN
View table
Table 5. Test accuracy of MAML and GNN
Model 1 2 3 4 5 Average
MAML 95.8 95.6 95.9 95.7 96.0 95.8
GNN 97.0 96.9 97.1 97.4 97.0 97.1

Table 6. Test accuracy of cross-domain detection tasks A to B and B to A
View table
Table 6. Test accuracy of cross-domain detection tasks A to B and B to A
Name 1 2 3 4 5 Average
A to B 96.7 96.8 96.6 96.6 96.8 96.7
B to A 96.4 96.1 96.7 96.3 96.4 96.4

Tools

Get Citation

Copy Citation Text

Huaian Yi, Runji Fang, Aihua Shu, Enhui Lu. Milling Surface Roughness Measurement Under Few-Shot Problem[J]. Laser & Optoelectronics Progress, 2022, 59(23): 2324001

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Optics at Surfaces

Received: Mar. 2, 2022

Accepted: Jun. 14, 2022

Published Online: Nov. 28, 2022

The Author Email: Huaian Yi (yihuaian@126.com)

DOI:10.3788/LOP2022059.2324001

Topics

laser devices and laser physics

Lasers and Laser Optics

laser manufacturing

Instrumentation, Measurement and Metrology