Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Types of defects to be detected.

(a) gasket pleat; (b) gasket missing;(c) dark point;(d) scratch.

More »

Fig 1 Expand

Fig 2.

The structure of the bottle cap to be detected.

(a) 3D schematic of the bottle cap; (b) commonly used high-angle ring light detection method; (c) the image brightness of the side wall region of the cap is dark, and the image of the inner circle region also has the problem of uneven background; (d) the image brightness of the side wall region is dark, but the brightness of the inner circle region is overexposed.

More »

Fig 2 Expand

Fig 3.

The detection system of LALDA.

(a) system schematic; (b) basic principles of low-angle incident light illumination; (c)(d) images of bottle caps in different colors captured by the system.

More »

Fig 3 Expand

Fig 4.

Algorithm flow.

More »

Fig 4 Expand

Fig 5.

Direct grayscale operation of different objects to be detected, as well as conversion to HSV color space: (a) the conversion of blue bottle cap; (b) the conversion of red bottle cap; (c) the conversion of pink bottle cap.

More »

Fig 5 Expand

Fig 6.

Different colors and H-channel images after HSV.

More »

Fig 6 Expand

Fig 7.

Sample with dark point and the V-channel in the side wall region.

More »

Fig 7 Expand

Fig 8.

Sample with dark point and the V-channel in the inner circle region.

More »

Fig 8 Expand

Fig 9.

Sample with scratch and the V-channel in the inner circle region.

More »

Fig 9 Expand

Fig 10.

Sample without defect and sample with gasket missing in the inner circle region.

More »

Fig 10 Expand

Fig 11.

Sample without defect and sample with gasket pleat in the inner circle region.

More »

Fig 11 Expand

Fig 12.

The actual built detection system based on LALDA.

(a) the main components of the detection system; (b), (c), (d) imaging results of the light source at different working distances (d1, d2, d3).

More »

Fig 12 Expand

Fig 13.

Comparison of different light intensities in the LASA.

(a) the light source controller is set to 40;(b) the light source controller is set to 70; (c)the light source controller is set to 140; (d)the light source controller is set to 200.

More »

Fig 13 Expand

Fig 14.

Comparison of traditional detection systems.

(a) the High-angle Light Detection Method;(b)the Coaxial lighting mode with the light source controller set to 90 degrees;(c) the High-angle Light Detection Method;(b)the Coaxial lighting mode with the light source controller set to 140 degrees.

More »

Fig 14 Expand

Fig 15.

Detection of dark points and color mixing in the interior of cap.

More »

Fig 15 Expand

Fig 16.

Detection of scratch in the inner circle region.

More »

Fig 16 Expand

Fig 17.

Detection of gasket missing in the inner circle region.

More »

Fig 17 Expand

Fig 18.

Detection of gasket pleat in the inner circle region.

More »

Fig 18 Expand

Fig 19.

Detection of different defects in bottle caps of different sizes and colors.

More »

Fig 19 Expand

Fig 20.

Comparison of results: Small black spots and scratches, as well as the shape, direction, and curvature of the scratches, have a certain randomness, leading to missed detections.

More »

Fig 20 Expand

Table 1.

Comparison of detection accuracy of different detection methods.

More »

Table 1 Expand