The Application Of Genetic Algorithm In GPR Prestack Migration

At present a serial of seismic data processing methods can be directly used in ground penetrating radar (GPR) multi-offset data processing. However, since the target in GPR exploration is usually shallow with respect to offset, conventional velocity analysis and normal moveout (NMO) correction are not suitable.The accuracy of migration image of ground-penetrating radar data is strongly related to the accuracy of permittivity distribution determined from multi-offset data. This paper proposes a migration velocity analysis method based on a global optimization method using genetic algorithm (GA). Under the assumptions that the media are piece wise, and that the permittivity of each piece can be expressed as a polynomial with limited terms, all the coefficients of the permittivity function of each piece, which maximize an objective function, are determined by genetic algorithm piece by piece.This paper also discuss a reverse-time prestack migration method using Maxwell's equations The finite-difference time-domain (FDTD) scheme for wave field simulation is used in both forward simulations and prestack migrations.A synthetic and a field data migration velocity analysis example including conventional velocity analysis result are given to illustrate the validation of the proposed method. The synthetic example shows that the proposed migration velocity analysis method can determine accurately the distribution of permittivity and that the migration stack profile fits the true model. The field migration stack profile is high quality.