Automatic velocity picking by simulated annealing

Conclusions

Simulated annealing is an effective global optimization method to cope with the nonlinear velocity picking problem. We cast two contradictory objective functions into a single objective function by a weighted-sum criterion. The weights for each function are chosen according to their importance. Experiments show that including prior information into the initialization is important both speed and convergence. The results demostrate the robustness of the algorithm.