How accurately do different computer-based texture characterization methods predict material surface coarseness? A guideline for effective online inspection
Sajjad Ghodrati, Saeideh Gorji Kandi, and Mohsen Mohseni
Sajjad Ghodrati, Saeideh Gorji Kandi, and Mohsen Mohseni, "How accurately do different computer-based texture characterization methods predict material surface coarseness? A guideline for effective online inspection," J. Opt. Soc. Am. A 35, 712-725 (2018)
The growing industrialization has emphasized the need for high-performance computer-based inspection methods. Here, we investigated the performance of four major computer-based texture characterization methods in the prediction of visually perceived and actual surface coarseness of real materials. Gray level co-occurrence matrix (GLCM), distance-dependent edge frequency (DDEF), fractal dimension (FD), and histogram skewness (SK) were used as the methods. A novel collection of real materials consisting of 20 sandpapers with high, medium, and low coarseness levels was employed. The results revealed that at high coarseness level the most precise prediction of actual surface coarseness was made by GLCM and SK, while in the prediction of visual coarseness all the methods worked similarly effectively. Perfect correlations were observed between GLCM, FD, and SK at visual and also actual coarseness at medium coarseness level. At low coarseness level, SK and DDEF acceptably predicted visual and actual coarseness, respectively. The image resolution impact on performance of the computer-based methods was found to be substantial. Results of the research present a guideline for choosing the best computer-based method as a viable substitute for the human observer in online inspections of materials’ texture.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Grit Values and Average Particle Diameters of the 20 Different Sandpapersa
High Coarseness Level
Medium Coarseness Level
Low Coarseness Level
Grit Value
Average Particle Diameter
Grit Value
Average Particle Diameter
Grit Value
Average Particle Diameter
P 24
764 μm
P 180
82 μm
P 800
21.8 μm
P 40
425 μm
P 220
68 μm
P 1000
18.3 μm
P 60
269 μm
P 280
52.2 μm
P 1200
15.3 μm
P 80
201 μm
P 320
46.2 μm
P 1500
12.6 μm
P 100
162 μm
P 400
35 μm
P 2000
10.3 μm
P 120
125 μm
P 600
25.8 μm
P 2500
8.4 μm
P 150
100 μm
–
–
P 5000
5 μm
The stimuli were classified into three coarseness categories based on their average abrasive particle diameters.
Table 2.
Best Correlation Coefficients between the GLCM Features and Actual Surface Coarseness and Also Visual Texture Coarseness for the Three Groups of Sandpapersa
The resolutions at which the best correlations were acquired have also been reported in the table.
Insignificant correlation ().
Table 3.
Best Correlation Coefficients between the DDEF and the Actual Surface Coarseness and Also the Visual Texture Coarseness for the Three Groups of Sandpapersa
High Coarseness Level
Medium Coarseness Level
Low Coarseness Level
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Edge Frequency
0.9286 (96 dpi)
0.9643 (96 dpi)
0.9429 (96 dpi)
0.9429 (96 dpi)
0.9643 (1200 dpi)
0.8571 (600 dpi)
The resolutions at which the best correlations were acquired have also been reported in the table.
Table 4.
Best Correlation Coefficients between the Box Counting and the Differential Box Counting Fractal Dimensions and Actual Surface Coarseness and Also Visual Texture Coarseness for the Three Groups of Sandpapersa
The resolutions at which the best correlations were acquired have also been reported in the table.
Insignificant correlation ().
Table 5.
Best Correlation Coefficients between the Histogram Skewness and Actual Surface Coarseness and Also Visual Texture Coarseness for the Three Groups of Sandpapersa
High Coarseness Level
Medium Coarseness Level
Low Coarseness Level
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Histogram Skewness
1 (1200 dpi)
0.9643 (1200 dpi)
1 (800 dpi)
1 (800 dpi)
0.8571 (1200 dpi)
0.9286 (800 dpi)
The resolutions at which the best correlations were acquired have also been reported.
Tables (5)
Table 1.
Grit Values and Average Particle Diameters of the 20 Different Sandpapersa
High Coarseness Level
Medium Coarseness Level
Low Coarseness Level
Grit Value
Average Particle Diameter
Grit Value
Average Particle Diameter
Grit Value
Average Particle Diameter
P 24
764 μm
P 180
82 μm
P 800
21.8 μm
P 40
425 μm
P 220
68 μm
P 1000
18.3 μm
P 60
269 μm
P 280
52.2 μm
P 1200
15.3 μm
P 80
201 μm
P 320
46.2 μm
P 1500
12.6 μm
P 100
162 μm
P 400
35 μm
P 2000
10.3 μm
P 120
125 μm
P 600
25.8 μm
P 2500
8.4 μm
P 150
100 μm
–
–
P 5000
5 μm
The stimuli were classified into three coarseness categories based on their average abrasive particle diameters.
Table 2.
Best Correlation Coefficients between the GLCM Features and Actual Surface Coarseness and Also Visual Texture Coarseness for the Three Groups of Sandpapersa
The resolutions at which the best correlations were acquired have also been reported in the table.
Insignificant correlation ().
Table 3.
Best Correlation Coefficients between the DDEF and the Actual Surface Coarseness and Also the Visual Texture Coarseness for the Three Groups of Sandpapersa
High Coarseness Level
Medium Coarseness Level
Low Coarseness Level
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Edge Frequency
0.9286 (96 dpi)
0.9643 (96 dpi)
0.9429 (96 dpi)
0.9429 (96 dpi)
0.9643 (1200 dpi)
0.8571 (600 dpi)
The resolutions at which the best correlations were acquired have also been reported in the table.
Table 4.
Best Correlation Coefficients between the Box Counting and the Differential Box Counting Fractal Dimensions and Actual Surface Coarseness and Also Visual Texture Coarseness for the Three Groups of Sandpapersa
The resolutions at which the best correlations were acquired have also been reported in the table.
Insignificant correlation ().
Table 5.
Best Correlation Coefficients between the Histogram Skewness and Actual Surface Coarseness and Also Visual Texture Coarseness for the Three Groups of Sandpapersa
High Coarseness Level
Medium Coarseness Level
Low Coarseness Level
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Actual Surface Coarseness
Visual Texture Coarseness
Histogram Skewness
1 (1200 dpi)
0.9643 (1200 dpi)
1 (800 dpi)
1 (800 dpi)
0.8571 (1200 dpi)
0.9286 (800 dpi)
The resolutions at which the best correlations were acquired have also been reported.