Centroid calculation provides a means of eliminating translation problems, which is useful for automatic target recognition. A neural network implementation of centroid calculation is described that uses a spatial filter and a Hopfield network to determine the centroid location of an object. Spatial filtering of a segmented window creates a result whose peak value occurs at the centroid of the input data set. A Hopfield network then finds the location of this peak and hence gives the location of the centroid. Hardware implementations of the networks are described and simulation results are provided.