In the unmanned machining, one of the most essential issue is the tool management system which includes controlling, identification, presetting and monitoring of cutting tools. Especially the monitoring of cutting tools may be the heart of TMS. During machining tool fracture causes workpieces demaged and downtime of machine tool increased. And tool wear produces the machining error as much as wear amounts. Consequently it is important to detect the condition of used tools and carry out an appropriate action.
In this paper the algorithm which can determine the tool condition is presented. The old tool image is taken by CCD camera and analysed quantitatively and qualitatively with image processing to get the information of rake and flank face of the tool. And then the modes of the tool fracture, crater, chipping and flank wear are classified using pattern recognition. The experimental results performed in the NC turning machine have proved the effectiveness of the system proposed.