Development of machine vision positioning system based on neural network

To resolve the localization problem of Industrial robot vision position product line, a fast and accurate algorithm of machine vision positioning is proposed based on radial basis function (RBF) neural network combined with improved Canny operator. This algorithm extracts the image feature edge with Canny Operator as a feature vector, learn the target image features, get RBF classifier and use it on the target location of the background image. Simulation results show that this method is applicable to the situation of high speed and precision of on-line machine vision position.