Abstract
This paper proposes an energy-minimizing level set method for defect detection in product surface, which consists of image segmentation module, image feature extraction module and product defect detection module. This new method, embedding energy function into the classic level set algorithm, segments internal and external areas of product image with minimized energy consumption, ensuring that the gradient of the level set function is in the direction of the minimum point so that the evolution process is closer to the zero level set. 15-dimentional features including zero-crossing rate and image entropy are used in the process of contour detection. Results show that the method is highly accurate and effective.
Similar content being viewed by others
References
Xiaocui, Y., Lushen, W., & Huawei, C. (2014). Improved image preprocessing algorithm for rail surface defects detection. Journal of Computer-Aided Design & Computer Graphics, 26(5), 800–805. (in Chinese).
Zhang, J., Ye, Y., & Xie, Y. (2014). Optoelectronic inspection of defects for metal cylindrical workpieces. Optics and Precision Engineering, 22(7), 1871–1876. (in Chinese).
Li, Q., Liang, Z., & Huang, Y. (2014). Sparseness representation model for defect detection and its application. Journal of Computer Research and Development, 51(9), 1929–1935. (in Chinese).
Wu, G., Sun, X., Pu, J., & Zhang, H. (2014). A novel surface defect detection method of cold rolled strips based on artificial immune system. In 2014 IEEE international conference on information and automation (ICIA) 28–30 July 2014, pp. 279–283.
Deschamps, T., & Cohen, L. D. (2002). Fast extraction of tubular and tree 3D surfaces with front propagation methods. In Proceedings of international conference on pattern recognition, pp. 731–734.
Cao, Z., Pan, S., Li, R., Balachandran, R., Fitzpatrick, M. J., Chapman, W. C., & Dawant, B. M. (2003). Registration of medical images using an interpolated closest point transform: method and validation. In: Proceedings of SPIE international society for optical engineering medical imaging, pp. 325–333.
Halim, S. A., Ibrahim, A., & Jayes, M. I. (2013). Weld defect features extraction on digital radiographic image using Chan–Vese model. In IEEE 9th international colloquium on signal processing and its applications, pp. 8–10.
Acknowledgements
The authors acknowledge the National Natural Science Foundation of China (Award Number: 51405435).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hu, K., Zhang, S. & Zhao, X. An Energy-Minimizing Level Set Method for Defect Detection. Wireless Pers Commun 102, 3545–3555 (2018). https://doi.org/10.1007/s11277-018-5390-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11277-018-5390-5