In this thesis, we present an optimal illumination condition and its model for extracting the tire information mark on the sidewall of a tire. Since the appearance of tire marks is the same as its background, a primary feature to extract tire marks is a roughness. Generally the roughness of tire mark is different from that of its background; the surface of the mark is smoother than the other parts. The light incident to the tire surface is reflected differently according to the roughness: the smoother surface has the higher intensity. Using this property we obtain the parameter values in Torrance-Sparrow reflectance model. To analyze the reflectance characteristics of the tire surface we measure the intensity while varying the incident angle of a light source and a viewing direction of CCD camera about the surface normal in turn. From the experimental results we find that the optimal illumination condition for extracting the tire information marks is when the incident angle of a light source is in accordance with viewing direction of a camera. We also develop an efficient reflectance ratio-based operator to extract the information mark.
