Application of an Improved Threshold Segmentation Algorithm in Lens Defect Detection

Yu Cao; Chuanpeng Xu

doi:10.3788/LOP202158.1610007

Laser & Optoelectronics Progress, Volume. 58, Issue 16, 1610007(2021)

Application of an Improved Threshold Segmentation Algorithm in Lens Defect Detection

Yu Cao^** and Chuanpeng Xu^*

Author Affiliations

School of Automation, Harbin University of Science and Technology, Harbin, Heilongjiang 150080, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(8)

Fig. 1. Ideal model of sub-pixel edge detection. (a) Original edge image model; (b) rotated edge image model

Download full size

Fig. 2. Curve inflection point diagram under different a and b values

Download full size

Fig. 3. Diagram of processing results under different a and b values. (a) a=2, b=15; (b) a=8, b=5; (c) a=4, b=10

Download full size

Fig. 4. Comparison of weight coefficients

Download full size

Fig. 5. Comparison of bubble threshold segmentation. (a) Original image of bubble; (b) Otsu segmentation; (c) standard PSO+Otsu segmentation; (d) improved PSO+Otsu segmentation; (e) improved two-dimensional maximum entropy threshold segmentation of genetic algorithm; (f) bubble edge

Download full size

Fig. 6. Comparison of scratch threshold segmentation. (a) Original image of scratches; (b) Otsu segmentation; (c) standard PSO+Otsu segmentation; (d) improved PSO+Otsu segmentation; (e) improved two-dimensional maximum entropy threshold segmentation of genetic algorithm; (f) scratch edge

Download full size

Table 1. Comparison of image segmentation of optical lens bubbles
View table
Table 1. Comparison of image segmentation of optical lens bubbles
Image Algorithm Time /ms Optimal threshold Number of iterations
Bubble Otsu 189 132 216
Standard PSO+ Otsu 156 133 14
Improved PSO+ Otsu 132 135 11
Ref. [16] method 147 135 23

Table 2. Comparison of image segmentation of optical lens scratches
View table
Table 2. Comparison of image segmentation of optical lens scratches
Image Algorithm Time /ms Optimal threshold Number of iterations
Scratches Otsu 196 141 216
Standard PSO+ Otsu 168 142 15
Improved PSO+ Otsu 141 143 10
Ref. [16] method 151 143 22

Tools

Get Citation

Copy Citation Text

Yu Cao, Chuanpeng Xu. Application of an Improved Threshold Segmentation Algorithm in Lens Defect Detection[J]. Laser & Optoelectronics Progress, 2021, 58(16): 1610007

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites