The surface quality of the component produced mainly depends on the condition of the grinding wheel. In this paper, the working surface condition of the grinding wheel is assessed by the texture analysis methods. The grinding wheel images are captured at different time intervals using a CCD camera. Then, the captured images are pre-processed using histogram equalisation to reduce the effect of non-uniform illumination. After pre-processing, Fourier power spectrum based texture parameters are evaluated and used for discriminating the grinding wheel condition. The evaluated parameters identify the condition of the grinding wheel efficiently, which in turn provide an opportunity for online condition monitoring of grinding wheel using machine vision. Copyright © 2010 Inderscience Enterprises Ltd.