Surface texture indicators of tool wear - A machine vision approach

There has been much research on the automated monitoring of cutting tool we ar. This research has tended to focus on three main areas that attempt to q uantify the cutting tool condition: monitoring of specific machine tool par ameters in order to infer tool condition, direct observations made on the c utting tool; and measurements taken from the chips produced by the tool. Ho wever, considerably less work has been performed on the development of surf ace texture sensors that provide information on the condition of the tool e mployed in machining the surface. A preliminary experimental study is prese nted for accomplishing this texture analysis using a machine vision-based s ensor system. In particular, an investigation of the condition of a two-flu te end mill used in a standard face milling operation is presented. The deg ree of tool wear is estimated by extracting three parameters from video cam era images of the machined surface. The performance of three image-processi ng algorithms, in estimating the tool condition, is presented: analysis of the intensity histogram; image frequency domain content; and spatial domain surface texture.