Investigation of flood embankment seepage using logistic regression with ICA

Abstract

This article analyzes the application of logistic regression with the use of ICA to detect leakages and leakages in flood embankments. As a result of the study, a model analysis was carried out where electrical impedance tomography tests were performed for one and for two rows of electrodes. For the results obtained, it was possible to verify the evaluation of the usefulness of the algorithms in the reconstruction of images derived from electrical impedance tomography. The proposed solution is a method that is non-invasive, and thus the embankment does not have to be destroyed for testing. Additionally, it should be emphasised that it is suitable for spatial imaging. The presented solution is a non-invasive method for assessing the structure of embankments, which is well suited for spatial imaging. The implemented algorithm shows promise in image reconstruction.