Texture analysis for grinding wheel wear assessment using machine vision

The surface quality of the components produced in a grinding process mainly depends on the grinding wheel topography. To maintain the quality of the component, the condition of the grinding wheel should be monitored and the periodical dressing has to be done to retain the shape and the sharpness of the cutting edges. In the present paper, the machine-vision-based texture analysis methods are introduced to discriminate the surface condition of the grinding wheel. The grinding wheel images are acquired using a charge coupled device (CCD) camera at different intervals of time during the process and the texture of the grinding wheel image is analysed using appropriate statistical and fractal methods. The variation of the parameters with the condition of the grinding wheel surface is analysed and presented. The evaluated parameters are clearly discriminating the texture of the grinding wheel at every stage. The methods presented in this paper can be used to assess the grinding wheel condition effectively. © IMechE 2007.