Deep Learning based Porosity Inversion from Seismic Attributes

For many applications in earth sciences and engineering, such as hydrocarbon reservoir characterization and geothermal energy generation, subsurface porosity is crucial. Many seismic factors, including Rayleigh waves, P, SH, SV, Love, and so on, have an impact on porosity. Numerous other requirements, including incoming seismic amplitude, transmitted amplitude, reflected amplitude, and reflection coefficients, are also met by these signals. The goal of this study is to forecast porosity through the application of Bagging and Deep Neural Net classification techniques along with the previously described multicriteria dataset analysis.