For the demands of wear on-line monitoring for mechanical equipment, an on-line oil monitoring system based on microscopic image analysis is constructed.According to the characteristic of system light route, the image of wear debris is converted into gray image based on its color feature, and the wear debris object is extracted by subtracting the background image from the wear debris image. The classifier for two kinds of wear debris is designed based on the least square support vector machines, and the parameters of this model are optimized by Particle Swarm Optimization(PSO) algorithm. Based on this classifier, an integrative wear debris classifier is designed according to the wear debris recognition system. The performance and recognition precision of this system are tested by the ferrography technology. The result shows that the recognition precision of this system is as high as 95%,which can meet the demand of wear debris on-line monitoring.