Noise suppression method based on multi-scale Dilated Convolution Network in desert seismic data

Yue Li, Yuying Wang, Ning Wu

Abstract: Highlights•The model can effectively remove noise in desert seismic records, recover the seismic events and enhance their continuity.•Combine multi-scale ideas with neural networks to extract and learn more abundant information.•The use of dilated convolutions is conducive to capture as much feature information from the noisy records.