Welding Defect Classification of Ultrasonic Detection Based on PCA and KNN

Abstract:

Aiming to problems of welding defect classification in the ultrasonic detection, according to the characteristics of welding defect, a classification method based on PCA and KNN is proposed in order to solve the problem of ultrasonic testing signal feature extraction and defect recognition. For the impact of the redundant attributes of feature extraction on the classification accuracy, in this paper, an ultrasonic flaw feature extraction algorithm based on PCA is proposed. Ultrasonic flaw intelligent classification is always a difficult problem in NDT. KNN algorithm is proposed to classify the different defects. Compared with BP algorithm, experiment results show that the model based on PCA and KNN can get stable classification results, high accuracy, and can effectively improve the classification efficiency.

• Periodical: Applied Mechanics and Materials (Volumes 380-384)
• Pages: 902-906
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.380-384.902
• Online since: August 2013
• Authors: Ai Ling Qi, Jing Fang Wang, Frank Wang, Unekwu Idachaba, Gbola Akanmu
• Keywords: Defect Classification, K-Nearest Neighbor (KNN), Principal Component Analysis (PCA), Ultrasonic Signal

References

[1] CHSNDT, Ultrasonic Detection. Beijing, Mechanical Industry Press, (2010).

[2] L. Han, Artificial Neural Network Tutorial. Beijing, Beijing University of Posts and Telecommunications Publishing House, (2006).

[3] J.B. Ma, J. Li, G.F. Teng, F. Wang, Y. Zhao, The Comparison Studies on the Algorithm of KNN and SVM for Chinese Text Automatic Classification, Journal of Agricultural University of Hebei, vol. 31(3), pp.120-123, (2008).

[4] Y. Sun, S.P. Lv, X.K. Wang, K-Nearest Algorithm Based on Learning Structure, Computer Science, vol. 34(12), pp.184-186, (2007).

[5] Y.M. Lin, W.D. Zhu, Study on the Application of Fuzzy KNN to Text Categorization, Computer Applications and Software, vol. 25 (9), pp.185-187, (2008).

[6] M. Zhang, T.S. Li, S.Y. Zhong, Implementation of Principal Component Analysis method (PCA) Based on Matlab, Journal of Guangxi University(Nat Sci Ed), vol. 30(sup), pp.74-77, (2005).

[7] Q. Niu, Z.X. Wang, D. Chen, S.X. Xia, Study on Web Document Classification Based on KNN, Computer Applications and Software, vol. 24(10), pp.210-211, (2008).

[8] Y.D. Liu, H.M. Niu, KNN Classification Algorithm Based on K-nearest neighbor Graph for Small Sample, Computer Engineering, vol. 37(9), pp.198-200, (2011).

[9] N. Meng, B. Wang, Research on KNN in Cardiotocograph classification on PCA, Computer Engineering, vol. 46(7), pp.223-225, (2010).