Surface roughness is one of the key factors to evaluate the grinding quality for engineering ceramics. This study introduces a new method based on grayscale information of surface images, to assess and predict the roughness of ground ceramics rapidly and effectively. The investigation sets the functional parameters of digital acquisition device as Brightness 140, Contrast 42, Saturation 24 and Acutance 9. Afterwards, it selects the mean value and the mean square deviation to describe surface roughness, and some image processing techniques are adopted to reduce noises and enhance the images. Lastly, it gives the relation curves on Ra，Rz，Ry versus grayscale information, and concludes a direct proportion law between the grayscale information and ground surface roughness.