Research on Experiment and Technique of Material Identification for Debris of Engineering Equipment Based on Computer Vision

Abstract:

Based on micro-image, the material identification of iron and copper debris in working oil of engineering equipment is researched. By selecting four characteristic parameters of debris image, a debris recognition classifier is designed based on multi-SVM. The materials and types of debris can be fast identified after the model is trained and the identification accuracy is high, thus a new method for fault diagnosis of engineering equipment is provided.

• Periodical: Advanced Materials Research (Volumes 479-481)
• Pages: 1115-1118
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.479-481.1115
• Online since: February 2012
• Authors: Qun Zhang Tu, Ming Pan, Jian Xun Zhao, Da Zhen Su, Guo Tao Wang
• Keywords: Image Processing, Material Identification, Support Vector Machine (SVM)

References

[1] Zuo HF: The Engine Wear Condition Monitoring and Fault Diagnose Technique (Aviation Industry Press, Beijing 1995).In Chinese

[2] Tu QZ, Zuo HF et.al: Journal of Nanjing University of Aeronautics and Astronautics vol. 10(2003), pp.506-509. In Chinese

[3] Yu ZH, Wang R: Modern Manufacturing Engineering vol.1(2010), pp.77-80. In Chinese

[4] Nello C, John S-T: Introduction to SVM (Publishing House of Electronic Industry, Beijing 2004).In Chinese

[5] Lv ZY, Yan XP: Lubrication Engineering vol.5(2005), pp.34-37. In Chinese