next up previous [pdf]

Next: Bibliography Up: Almomin: WEMVA Previous: Modified correlation-based WEMVA

Discussion and Conclusions

By examining the results, we see that, when provided with the true depth, the correlation-based WEMVA produces excellent results with no cycle-skipping or edge effects. When the apparent depth was provided, the results start to have some errors. However, these errors are to be expected due to the velocity-depth ambiguity. Therefore, the gradients will flatten the gathers but not necessarily resolve the correct depth (Zhang and Biondi, 2011). However, the results are still satisfactory, since the gradient is pointing toward the right direction.

Using the modified gradient has several advantages, such as handling multiple events. However, I need to further test the approximations made in the modified gradient. Leeuwen (2010) showed that it is possible to further improve the objective function by multiplying with the proper weights and minimizing the correlation function instead of the lags, which will eliminate any picking. Finally, using correlation functions over extended images (Yang and Sava, 2009) could help provide better results.