A neural network for probabilistic edge labelling trained with a step edge model

This paper presents a robust neural network edge labelling strategy in which a network is trained with data from an imaging model of an ideal step edge. In addition to the Sobel operator, the authors employ preprocessing steps on image data to exploit the known invariances due to lighting variation and rotation and so reduce the complexity of the mapping which the network has to learn. The composition of the training set to achieve labelling of the image lattice with Baysian posterior probabilities is described. Results are shown for real images and comparison made with the Canny edge detector: the effects of adding zero-mean Gaussian noise are also shown. To elucidate the roles of the Sobel operator and the network a probabilistic Sobel labelling strategy has been derived-its results are inferior to those of the neural network