Multiresolution data fusion algorithm for aerospace applications

Neural network based data fusion techniques are becoming increasingly popular for quantifying corrosion in aircraft lap joints because of their accuracy and ability to learn the mapping between measurements and the corresponding wall thickness. However, this approach to data fusion is slow and less robust because of high input dimensionality. This paper proposes a multiresolution based data fusion algorithm to improve the performance of corrosion quantification in aircraft lap joints. The proposed algorithm reduces the dimensionality of the input and improves the robustness of the network by separating fusion and corrosion characterization operations. Initial results validate these claims and indicate the feasibility of the algorithm.