3D-measurement of geometrical shapes by photogrammetry and neural networks

A method is introduced which couples the classical estimation of 3D-coordinates with processing in an artificial neural network (ANN). The ANN is used to reduce the random and systematic errors of the measurement values by a-priori knowledge. The calculated geometrical shape is more precise than the results obtained with other methods or needs fewer measurement values. To calculate the weights suitable algorithms are used. It is possible to measure special dimensions of parts of measurement objects.