Vision Servoing Method Based on Texture Analyzing

By analyzing the relationship between the camera motion and the coarseness changing of the texture image, a vision servoing method based on texture coarseness analyzing is given in this paper. The theory based on geography feature calculation is not effective in the environment full of texture but short of geography features. Texture feature extraction corresponding to the camera translation, pitching and yawing is talked. The result of half-physical robot station keeping experiment shows that the changing of texture features can provide enough motion information for the servoing task of the robot.