The issue of strong noise has increasingly become a bottleneck restricting the precision and application space of electromagnetic exploration methods.Noise suppression and extraction of effective electromagnetic respo...The issue of strong noise has increasingly become a bottleneck restricting the precision and application space of electromagnetic exploration methods.Noise suppression and extraction of effective electromagnetic response information under a strong noise background is a crucial scientific task to be addressed.To solve the noise suppression problem of the controlled-source electromagnetic method in strong interference areas,we propose an approach based on complex-plane 2D k-means clustering for data processing.Based on the stability of the controlled-source signal response,clustering analysis is applied to classify the spectra of different sources and noises in multiple time segments.By identifying the power spectra with controlled-source characteristics,it helps to improve the quality of the controlled-source response extraction.This paper presents the principle and workflow of the proposed algorithm,and demonstrates feasibility and effectiveness of the new algorithm through synthetic and real data examples.The results show that,compared with the conventional Robust denoising method,the clustering algorithm has a stronger suppression effect on common noise,can identify high-quality signals,and improve the preprocessing data quality of the controlledsource electromagnetic method.展开更多
In the study of spiral galaxy morphology,spiral arm structures are valuable for intuitively reflecting active physical and chemical processes within galaxies.However,long-term scarcity of high-quality one-,three-,and ...In the study of spiral galaxy morphology,spiral arm structures are valuable for intuitively reflecting active physical and chemical processes within galaxies.However,long-term scarcity of high-quality one-,three-,and four-armed galaxy samples has limited deep learning model performance.To address this,this study developed a spiral galaxy data simulation program with a three-stage workflow:first,screening highly reliable training samples;second,selecting the best-performing Imagen architecture as the generative model after comparing nine mainstream ones;finally,training Imagen to generate an open data set of 9402 one-/three-armed galaxies,expanding the original sample size by 6 times.Multi-dimensional evaluations verified reliability and usability:Fréchet Inception Distance scores for N=1 and N=3 tasks were 6.05 and 9.13;the t-distributed Stochastic Neighbor Embedding showed generated data covered and expanded real data distribution;the Structural Similarity Index Measure confirmed no sample duplication.In downstream validation,data augmentation improved seven classification models'average accuracy by 8.7%(DenseNet peaked at 97%),and SHapley Additive exPlanations analysis showed model decisions focused on spiral arm topology.In conclusion,the program and data set support spiral galaxy morphology deep learning research and are publicly available at https://github.com/TuAstroAILab/AstroGS.展开更多
基金supported by the National Key Research and Development Program Project of China(Grant No.2023YFF0718003)the key research and development plan project of Yunnan Province(Grant No.202303AA080006).
文摘The issue of strong noise has increasingly become a bottleneck restricting the precision and application space of electromagnetic exploration methods.Noise suppression and extraction of effective electromagnetic response information under a strong noise background is a crucial scientific task to be addressed.To solve the noise suppression problem of the controlled-source electromagnetic method in strong interference areas,we propose an approach based on complex-plane 2D k-means clustering for data processing.Based on the stability of the controlled-source signal response,clustering analysis is applied to classify the spectra of different sources and noises in multiple time segments.By identifying the power spectra with controlled-source characteristics,it helps to improve the quality of the controlled-source response extraction.This paper presents the principle and workflow of the proposed algorithm,and demonstrates feasibility and effectiveness of the new algorithm through synthetic and real data examples.The results show that,compared with the conventional Robust denoising method,the clustering algorithm has a stronger suppression effect on common noise,can identify high-quality signals,and improve the preprocessing data quality of the controlledsource electromagnetic method.
基金supported by the National Natural Science Foundation of China(NSFC,grant No.U1731128)the support of the Association for Astronomy X A.I.(A3),funded by the Science and Education Integration Funding of the University of Chinese Academy of Sciences。
文摘In the study of spiral galaxy morphology,spiral arm structures are valuable for intuitively reflecting active physical and chemical processes within galaxies.However,long-term scarcity of high-quality one-,three-,and four-armed galaxy samples has limited deep learning model performance.To address this,this study developed a spiral galaxy data simulation program with a three-stage workflow:first,screening highly reliable training samples;second,selecting the best-performing Imagen architecture as the generative model after comparing nine mainstream ones;finally,training Imagen to generate an open data set of 9402 one-/three-armed galaxies,expanding the original sample size by 6 times.Multi-dimensional evaluations verified reliability and usability:Fréchet Inception Distance scores for N=1 and N=3 tasks were 6.05 and 9.13;the t-distributed Stochastic Neighbor Embedding showed generated data covered and expanded real data distribution;the Structural Similarity Index Measure confirmed no sample duplication.In downstream validation,data augmentation improved seven classification models'average accuracy by 8.7%(DenseNet peaked at 97%),and SHapley Additive exPlanations analysis showed model decisions focused on spiral arm topology.In conclusion,the program and data set support spiral galaxy morphology deep learning research and are publicly available at https://github.com/TuAstroAILab/AstroGS.
