This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Faculty of Architecture, Civil Engineering and Geodesy, University of Banja Luka, Banja Luka, Bosnia and Herzegovina
Faculty of Architecture, Civil Engineering and Geodesy, University of Banja Luka, Banja Luka, Bosnia and Herzegovina
Faculty of Civil Engineering, University of Belgrade, Belgrade, Serbia
Faculty of Civil Engineering, University of Belgrade, Belgrade, Serbia
The transition from ellipsoidal geodetic coordinates to rectangular 3D coordinates is quite simple, while the reverse procedure is somehow more complex due to the mathematical relationship between the ellipsoidal width and 3D coordinates. Until now, several methods for solving this problem have been defined and described in geodetic literature. In this paper, the possibility of applying a backpropagation algorithm based on a multilayer perceptron (Multilayer Perceptron – MLP) neural network for the transformation of rectangular 3D geodetic into ellipsoidal coordinates is analyzed. The applied MLP model is based on Bayesian regularization (BR). The adequacy of the model was verified by a robustness test and a cross-validation test. Based on the obtained results, it was concluded that the MLP neural network can be used for the transformation from rectangular 3D to ellipsoidal coordinates. Future research should analyze the possibility of applying this procedure to solve the problem of data transformation.
Coordinate Transformation, Multilayer Perceptron (MLP) Neural Network, Bayesian Regularization
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.
