BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals

Airborne transient electromagnetic (ATEM) surveys provide a fast, flexible approach for identifying conductive metal deposits across a variety of intricate terrains. Nonetheless, the secondary electromagnetic response signals captured by ATEM systems frequently suffer from numerous noise interferenc...

Full description

Saved in:
Bibliographic Details
Main Authors: Weijie Wang, Xuben Wang, Xiaodong Yu, Debiao Luo, Xinyue Liu, Kai Yang, Wen Yang, Xiaolan Yang, Ke Hu, Wenyi Hu
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/1/77
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850116914363236352
author Weijie Wang
Xuben Wang
Xiaodong Yu
Debiao Luo
Xinyue Liu
Kai Yang
Wen Yang
Xiaolan Yang
Ke Hu
Wenyi Hu
author_facet Weijie Wang
Xuben Wang
Xiaodong Yu
Debiao Luo
Xinyue Liu
Kai Yang
Wen Yang
Xiaolan Yang
Ke Hu
Wenyi Hu
author_sort Weijie Wang
collection DOAJ
description Airborne transient electromagnetic (ATEM) surveys provide a fast, flexible approach for identifying conductive metal deposits across a variety of intricate terrains. Nonetheless, the secondary electromagnetic response signals captured by ATEM systems frequently suffer from numerous noise interferences, which impede effective data processing and interpretation. Traditional denoising methods often fall short in addressing these complex noise backgrounds, leading to less-than-optimal signal extraction. To tackle this issue, a deep learning-based denoising network, called BA-ATEMNet, is introduced, using Bayesian learning alongside a multi-head self-attention mechanism to effectively denoise ATEM signals. The incorporation of a multi-head self-attention mechanism significantly enhances the feature extraction capabilities of the convolutional neural network, allowing for improved differentiation between signal and noise. Moreover, the combination of Bayesian learning with a weighted integration of prior knowledge and SNR enhances the model’s performance across varying noise levels, thereby increasing its adaptability to complex noise environments. Our experimental findings indicate that BA-ATEMNet surpasses other denoising models in both single and multiple noise conditions, achieving an average signal-to-noise ratio of 37.21 dB in multiple noise scenarios. This notable enhancement in SNR, compared to the next best model, which achieves an average SNR of 36.10 dB, holds substantial implications for ATEM-based mineral exploration and geological surveys.
format Article
id doaj-art-2e4888e0c2cd436c8bbd9abe8d9c7635
institution OA Journals
issn 1424-8220
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-2e4888e0c2cd436c8bbd9abe8d9c76352025-08-20T02:36:12ZengMDPI AGSensors1424-82202024-12-012517710.3390/s25010077BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic SignalsWeijie Wang0Xuben Wang1Xiaodong Yu2Debiao Luo3Xinyue Liu4Kai Yang5Wen Yang6Xiaolan Yang7Ke Hu8Wenyi Hu9School of Geophysics, Chengdu University of Technology, Chengdu 610059, ChinaSchool of Geophysics, Chengdu University of Technology, Chengdu 610059, ChinaSchool of Computer Science, Chengdu University, Chengdu 610106, ChinaInformation Network Center, Chengdu University, Chengdu 610106, ChinaInformation Network Center, Chengdu University, Chengdu 610106, ChinaSchool of Geophysics, Chengdu University of Technology, Chengdu 610059, ChinaInformation Network Center, Chengdu University, Chengdu 610106, ChinaInformation Network Center, Chengdu University, Chengdu 610106, ChinaInformation Network Center, Chengdu University, Chengdu 610106, ChinaInformation Network Center, Chengdu University, Chengdu 610106, ChinaAirborne transient electromagnetic (ATEM) surveys provide a fast, flexible approach for identifying conductive metal deposits across a variety of intricate terrains. Nonetheless, the secondary electromagnetic response signals captured by ATEM systems frequently suffer from numerous noise interferences, which impede effective data processing and interpretation. Traditional denoising methods often fall short in addressing these complex noise backgrounds, leading to less-than-optimal signal extraction. To tackle this issue, a deep learning-based denoising network, called BA-ATEMNet, is introduced, using Bayesian learning alongside a multi-head self-attention mechanism to effectively denoise ATEM signals. The incorporation of a multi-head self-attention mechanism significantly enhances the feature extraction capabilities of the convolutional neural network, allowing for improved differentiation between signal and noise. Moreover, the combination of Bayesian learning with a weighted integration of prior knowledge and SNR enhances the model’s performance across varying noise levels, thereby increasing its adaptability to complex noise environments. Our experimental findings indicate that BA-ATEMNet surpasses other denoising models in both single and multiple noise conditions, achieving an average signal-to-noise ratio of 37.21 dB in multiple noise scenarios. This notable enhancement in SNR, compared to the next best model, which achieves an average SNR of 36.10 dB, holds substantial implications for ATEM-based mineral exploration and geological surveys.https://www.mdpi.com/1424-8220/25/1/77deep learningairborne transient electromagnetic methodmulti-head self-attention mechanismBayesian learningsignal denoising
spellingShingle Weijie Wang
Xuben Wang
Xiaodong Yu
Debiao Luo
Xinyue Liu
Kai Yang
Wen Yang
Xiaolan Yang
Ke Hu
Wenyi Hu
BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals
Sensors
deep learning
airborne transient electromagnetic method
multi-head self-attention mechanism
Bayesian learning
signal denoising
title BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals
title_full BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals
title_fullStr BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals
title_full_unstemmed BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals
title_short BA-ATEMNet: Bayesian Learning and Multi-Head Self-Attention for Theoretical Denoising of Airborne Transient Electromagnetic Signals
title_sort ba atemnet bayesian learning and multi head self attention for theoretical denoising of airborne transient electromagnetic signals
topic deep learning
airborne transient electromagnetic method
multi-head self-attention mechanism
Bayesian learning
signal denoising
url https://www.mdpi.com/1424-8220/25/1/77
work_keys_str_mv AT weijiewang baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT xubenwang baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT xiaodongyu baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT debiaoluo baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT xinyueliu baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT kaiyang baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT wenyang baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT xiaolanyang baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT kehu baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals
AT wenyihu baatemnetbayesianlearningandmultiheadselfattentionfortheoreticaldenoisingofairbornetransientelectromagneticsignals