Open AccessSeismic reflections de-noising and recognition using Empirical Mode Decomposition and Continuous Wavelet Transformation
Author(s) -
Amjad Ali,
ShengChang Chen
Publication year - 2022
Publication title -
natural and applied sciences international journal
Language(s) - English
Resource type - Journals
ISSN - 2788-4619
DOI - 10.47264/idea.nasij/3.1.1
Subject(s) - hilbert–huang transform , wavelet , transformation (genetics) , noise (video) , signal (programming language) , computer science , filter (signal processing) , pattern recognition (psychology) , time domain , continuous wavelet transform , signal processing , gaussian noise , algorithm , artificial intelligence , speech recognition , wavelet transform , discrete wavelet transform , computer vision , digital signal processing , biochemistry , chemistry , computer hardware , image (mathematics) , gene , programming language
Current developments in signal processing are allowing for enhanced seismic illustrations and investigation of subsurface structures. Recently, Empirical Mode Decomposition (EMD) and Continuous Wavelet Transformation (CWT) have been introduced to extract various features from a time series dataset. In this investigation, seismic signal with 10% Gaussian noise is transformed into sub-signals by EMD analysis to improve the Signal-to-Noise Ratio (SNR). Then, CWT is implemented for each sub-signal to identify the exact locations of seismic reflections. The main objective of this study is to utilize the EMD as a noise filter in the time-domain and CWT to recognize the anomalous zone in each sub-signal. Based on the results of EMD and CWT, the true representation of a seismic signal with minimum noise in the time domain has been achieved. The successful integration of EMD and CWT is achieved in terms of the identification of true seismic reflections as localized anomalous zones at 0.8 sec, 1 sec, and 1.07 sec.