Study on the Parts Surface Defect Detection Method Based on Modifed SVM Algorithm

Abstract:

In this paper, we propose an efficient parts surface defect detection method using SVM algorithm, and particle swarm optimization is used to make parameters selection for SVM. The proposed parts surface defect detection systems is made up of six parts, and the main ideas of our method lie in that we exploit computer vision and machine learning in the research field of mechanical manufacturing and automation. We convert the parts surface defect detection problem to the classification problem, and the images of parts surface are used as testing samples. The SVM algorithm regards the classification problem as the constrained optimization problem. The classification accuracy is determined by the quality of parameters selection. Hence, particle swarm optimization is exploited to make parameters selection for SVM by defining two fitness functions. Afterwards, the best particle of the current population and the gbest is obtained. Utilizing the output from the particle swarm optimization then the parameters for SVM can be obtained. Finally, experiments are conducted based on a dataset with 563 samples, and experimental results illustrate that the proposed is quite effective for parts surface defect detection.

• Periodical: Applied Mechanics and Materials (Volumes 541-542)
• Pages: 1447-1451
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.541-542.1447
• Online since: March 2014
• Authors: Feng Ying Zhang, Rui Juan Liu
• Keywords: Optimization Problem, Particle Swarm Optimization (PSO), Parts Surface Defect, SVM, Vector

References

[1] Zhou Awei, Guo Junjie, Shao Wei, Multipose measurement of surface defects on rotary metal parts with a combined laser-and-camera sensor, OPTICAL ENGINEERING, 2013, 52(10), Article No. 104104.

DOI: 10.1117/1.oe.52.10.104104

[2] Zhou Awei, Guo Junjie, Shao Wei, Automated Detection of Surface Defects on Sphere Parts Using Laser and CDD Measurements, IEEE ICELIE/IES Industry Forum/37th Annual Conference of the IEEE Industrial-Electronics-Society (IECON), 2011 , pp.2666-2671.

DOI: 10.1109/iecon.2011.6119732

[3] Santoni A., Guo M. -C., Heuzey M. -C., Surface defects of TPO injected foam parts for automotive applications, International Polymer Processing, 2007, 22(2): 204-212.

DOI: 10.1515/ipp-2007-0012

[4] Zhang Guoshan, Wang Shiwei, Wang Yiming, LS-SVM Approximate Solution for Affine Nonlinear Systems with Partially Unknown Functions, Journal of Industrial and Management Optimization, 2014, 10(2): 621-636.

DOI: 10.3934/jimo.2014.10.621

[5] Zhou Ligang, Lai Kin Keung, Yen Jerome, Bankruptcy prediction using SVM models with a new approach to combine features selection and parameter optimisation, International Journal of Systems Science, 2014, 45(3): 241-253.

DOI: 10.1080/00207721.2012.720293

[6] Wang Yanqiu, Li Yang, Wang Qi, Computational identification of human long intergenic non-coding RNAs using a GA-SVM algorithm, GENE, 2014, 533(1): 94-99.

DOI: 10.1016/j.gene.2013.09.118

[7] Hou Chenping, Nie Feiping, Zhang Changshui, Multiple rank multi-linear SVM for matrix data classification, Pattern Recognition, 2014, 47(1): 454-469.

DOI: 10.1016/j.patcog.2013.07.002

[8] Chen Cheng-Hung, Liao Yen-Yun, Tribal particle swarm optimization for neurofuzzy inference systems and its prediction applications, COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION, 2014, 19(4): 914-929.

DOI: 10.1016/j.cnsns.2013.08.022

[9] Hu Yifan, Ding Yongsheng, Hao Kuangrong, An immune orthogonal learning particle swarm optimisation algorithm for routing recovery of wireless sensor networks with mobile sink, INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE, 2014, 45(3): 337-350.

DOI: 10.1080/00207721.2012.723053

[10] Shamsudin Supiah, Dan'azumi Salisu, Aris Azmi, Optimum combination of pond volume and outlet capacity of a stormwater detention pond using particle swarm optimization, URBAN WATER JOURNAL, 2014, 11(2): 127-136.

DOI: 10.1080/1573062x.2013.768680

[11] Chen Fafa, Tang Baoping, Song Tao, Multi-fault diagnosis study on roller bearing based on multi-kernel support vector machine with chaotic particle swarm optimization, MEASUREMENT, 2014, 47: 576-590.

DOI: 10.1016/j.measurement.2013.08.021