This paper describes a non-contact technique to assess the differences in surface characteristics of the ground components. The computer vision based system is used to analyze the pattern of scattered light from the surface to asses the surface roughness of the component. The ground specimens were manufactured using varying machining parameters. The images of the specimens are captured using a CCD camera. The image parameters based on the wavelet transform are evaluated. Then, the evaluated parameters along with the cutting parameters were used to train the artificial neural network to predict the surface roughness parameters Ra, which is measured using the stylus instrument. The comparison of stylus Ra and that predicted using ANN are presented and analyzed in this paper. Copyright© (2006) by the International Measurement Federation (IMEKO).