Limits...
A Novel Machine Vision System for the Inspection of Micro-Spray Nozzle.

Huang KY, Ye YT - Sensors (Basel) (2015)

Bottom Line: These texture features (contrast, entropy, energy), color features (mean and variance of gray level) and geometric features (distance variance, mean diameter and diameter ratio) were used in the classification procedures.A back-propagation neural network classifier was employed to detect the defects of micro-spray nozzles.The methodology presented herein effectively works for detecting micro-spray nozzle defects to an accuracy of 90.71%.

View Article: PubMed Central - PubMed

Affiliation: Department of Bio-Industrial Mechatronics Engineering, National Chung Hsing University, Tai-Chung 402, Taiwan. kuoyi@nchu.edu.tw.

ABSTRACT
In this study, we present an application of neural network and image processing techniques for detecting the defects of an internal micro-spray nozzle. The defect regions were segmented by Canny edge detection, a randomized algorithm for detecting circles and a circle inspection (CI) algorithm. The gray level co-occurrence matrix (GLCM) was further used to evaluate the texture features of the segmented region. These texture features (contrast, entropy, energy), color features (mean and variance of gray level) and geometric features (distance variance, mean diameter and diameter ratio) were used in the classification procedures. A back-propagation neural network classifier was employed to detect the defects of micro-spray nozzles. The methodology presented herein effectively works for detecting micro-spray nozzle defects to an accuracy of 90.71%.

No MeSH data available.


(a) Original image; (b) segmented binary image after the CI algorithm; (c) segmented image after the CI algorithm; (d) classification result.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541833&req=5

sensors-15-15326-f013: (a) Original image; (b) segmented binary image after the CI algorithm; (c) segmented image after the CI algorithm; (d) classification result.

Mentions: The difference between defects and non-defect regions is not obvious in the original image of a micro-spray nozzle, as shown in Figure 13a. Possible defect regions, including defects and non-defects, are segmented using the CI algorithm (parameters n = 6, T = 11 are selected after some experimentations) and image processing techniques (such as hole-filling, erosion, dilation, opening, closing and Canny edge operators), as indicated in Figure 13b. Figure 13c results using the AND logic operator for Figure 13a,b. Hence, the defects of micro-spray nozzles (Figure 13d) are detected according to the BPNN classification.


A Novel Machine Vision System for the Inspection of Micro-Spray Nozzle.

Huang KY, Ye YT - Sensors (Basel) (2015)

(a) Original image; (b) segmented binary image after the CI algorithm; (c) segmented image after the CI algorithm; (d) classification result.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541833&req=5

sensors-15-15326-f013: (a) Original image; (b) segmented binary image after the CI algorithm; (c) segmented image after the CI algorithm; (d) classification result.
Mentions: The difference between defects and non-defect regions is not obvious in the original image of a micro-spray nozzle, as shown in Figure 13a. Possible defect regions, including defects and non-defects, are segmented using the CI algorithm (parameters n = 6, T = 11 are selected after some experimentations) and image processing techniques (such as hole-filling, erosion, dilation, opening, closing and Canny edge operators), as indicated in Figure 13b. Figure 13c results using the AND logic operator for Figure 13a,b. Hence, the defects of micro-spray nozzles (Figure 13d) are detected according to the BPNN classification.

Bottom Line: These texture features (contrast, entropy, energy), color features (mean and variance of gray level) and geometric features (distance variance, mean diameter and diameter ratio) were used in the classification procedures.A back-propagation neural network classifier was employed to detect the defects of micro-spray nozzles.The methodology presented herein effectively works for detecting micro-spray nozzle defects to an accuracy of 90.71%.

View Article: PubMed Central - PubMed

Affiliation: Department of Bio-Industrial Mechatronics Engineering, National Chung Hsing University, Tai-Chung 402, Taiwan. kuoyi@nchu.edu.tw.

ABSTRACT
In this study, we present an application of neural network and image processing techniques for detecting the defects of an internal micro-spray nozzle. The defect regions were segmented by Canny edge detection, a randomized algorithm for detecting circles and a circle inspection (CI) algorithm. The gray level co-occurrence matrix (GLCM) was further used to evaluate the texture features of the segmented region. These texture features (contrast, entropy, energy), color features (mean and variance of gray level) and geometric features (distance variance, mean diameter and diameter ratio) were used in the classification procedures. A back-propagation neural network classifier was employed to detect the defects of micro-spray nozzles. The methodology presented herein effectively works for detecting micro-spray nozzle defects to an accuracy of 90.71%.

No MeSH data available.