科研之友
微信
新浪微博
Facebook
分享链接
摘要
Seismograms exhibit a good approximation of a geological structure. However, the images they show are generally contaminated by irrelevant information. The noise ground roll in these images can contribute significantly to the distortion of the data present in the desired information, due to the scattering of waves in deeper regions of geological layers. In this work, we used a method based on Haar and Daubechies wavelets applied in conjunction with artificial neural networks to reduce the noise ground roll. This type of noise is normally present in earth seismic images and it is similar to those found in oil reservoirs.
- 出版日期2015-2
