Abstract：The development of oil particle dynamic analysis system is described, and dynamic particle image detection is discussed. On the basis of analyzing degradation theory about dynamic particle image, simulation of degradation of standard debris atlas images is carried out according to the system experimentation parameters, and the particle extraction from degraded image is analyzed and resolved, then the variability of various parameters is studied. At last by means of selecting the parameters with steadiness of degradation, the oil dynamic particle image detection system is developed and comparing with conventional systems more detailed detection and classification are realized.
［1］Reintjes J,Tucker J E, et al. Application of LaserNet Fines to Mechanical Wear and Hydraulic Monitoring［C］//DSTO International Conference on Health and Usage Monitoring.Melboure, Australia, 2001:13-21.
［2］史明德,龙怀忠. 发散型微功耗光学流体污染检测传感器: 中国,99115011.2［P］.2003.
［3］吕植勇. 磨粒检测数字化方法的研究［D］. 武汉:武汉理工大学, 2006.
［4］Miller N A,Henderson J J.Quantifying sand particle shape complexity using a dynamic digital imaging technique［J］. Agronomy Journal, 2010,102(5) :1407-1414.
［5］武通海,邱辉鹏,吴教义,等. 图像可视在线铁谱传感器的图像数字化处理技术［J］.机械工程学报, 2008,44(9):83-87.
［6］辛登科,张玉杰,苏治果. 图像处理在粉末粒度在线检测系统中的应用［J］. 计算机工程与设计,2008,29(13) :3510-3512.
［7］王旭辉,郭光亚.二维匀速运动模糊图像恢复问题的研究［J］.计算机应用,2000, 20(10):25-28.
［8］赵琳,金伟其,黄有为,等. 离焦模糊图像超分辨力盲复原算法分析［J］.光学技术,2010,36(1):75-78.
［9］安德森 D P.磨粒图谱［M］. 金元生, 杨其明译. 北京:机械工业出版社, 1987.
2012, Vol. 33 
