Use of image processing to collect the time data of NC machining

To control the production schedule for mold making, it is important to collect the actual results of machining operations, yet machine operators seldom record the results. Some die and mold shops have constructed a system which interfaces with NC controllers and which identifies the status of the machining operation from signals. This method depends totally on the NC controller and requires higher installation cost. We propose a method that is independent of the NC controller. This method uses a low-cost PC camera and image processing technology; and it has one algorithm for tracking moving objects and another for detecting scene changes. To demonstrate the validity of the proposed method, we applied it to both attended and unattended NC machining.