{"title":"DSRL:基于离散小波变换和深度学习方法的 LAMOST 低分辨率恒星光谱自动分类方法","authors":"Hao Li, Qing Zhao, Chengkui Zhang, Chenzhou Cui, Dongwei Fan, Yuan Wang, Yarui Chen","doi":"10.1007/s10686-024-09940-0","DOIUrl":null,"url":null,"abstract":"<div><p>Automatic classification of stellar spectra contributes to the study of the structure and evolution of the Milky Way and star formation. Currently available methods exhibit unsatisfactory spectral classification accuracy. This study investigates a method called DSRL, which is primarily used for automated and accurate classification of LAMOST stellar spectra based on MK classification criteria. The method utilizes discrete wavelet transform to decompose the spectra into high-frequency and low-frequency information, and combines residual networks and long short-term memory networks to extract both high-frequency and low-frequency features. By introducing self-distillation (DSRL-1, DSRL-2, and DSRL-3), the classification accuracy is improved. DSRL-3 demonstrates superior performance across multiple metrics compared to existing methods. In both three-class(F ,G ,K) and ten-class(A0, A5, F0, F5, G0, G5, K0, K5, M0, M5) experiments, DSRL-3 achieves impressive accuracy, precision, recall, and F1-Score results. Specifically, the accuracy performance reaches 94.50% and 97.25%, precision performance reaches 94.52% and 97.29%, recall performance reaches 94.52% and 97.22%, and F1-Score performance reaches 94.52% and 97.23%. The results indicate the significant practical value of DSRL in the classification of LAMOST stellar spectra. To validate the model, we visualize it using randomly selected stellar spectral data. The results demonstrate its excellent application potential in stellar spectral classification.</p></div>","PeriodicalId":551,"journal":{"name":"Experimental Astronomy","volume":"57 3","pages":""},"PeriodicalIF":2.7000,"publicationDate":"2024-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"DSRL: A low-resolution stellar spectral of LAMOST automatic classification method based on discrete wavelet transform and deep learning methods\",\"authors\":\"Hao Li, Qing Zhao, Chengkui Zhang, Chenzhou Cui, Dongwei Fan, Yuan Wang, Yarui Chen\",\"doi\":\"10.1007/s10686-024-09940-0\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Automatic classification of stellar spectra contributes to the study of the structure and evolution of the Milky Way and star formation. Currently available methods exhibit unsatisfactory spectral classification accuracy. This study investigates a method called DSRL, which is primarily used for automated and accurate classification of LAMOST stellar spectra based on MK classification criteria. The method utilizes discrete wavelet transform to decompose the spectra into high-frequency and low-frequency information, and combines residual networks and long short-term memory networks to extract both high-frequency and low-frequency features. By introducing self-distillation (DSRL-1, DSRL-2, and DSRL-3), the classification accuracy is improved. DSRL-3 demonstrates superior performance across multiple metrics compared to existing methods. In both three-class(F ,G ,K) and ten-class(A0, A5, F0, F5, G0, G5, K0, K5, M0, M5) experiments, DSRL-3 achieves impressive accuracy, precision, recall, and F1-Score results. Specifically, the accuracy performance reaches 94.50% and 97.25%, precision performance reaches 94.52% and 97.29%, recall performance reaches 94.52% and 97.22%, and F1-Score performance reaches 94.52% and 97.23%. The results indicate the significant practical value of DSRL in the classification of LAMOST stellar spectra. To validate the model, we visualize it using randomly selected stellar spectral data. The results demonstrate its excellent application potential in stellar spectral classification.</p></div>\",\"PeriodicalId\":551,\"journal\":{\"name\":\"Experimental Astronomy\",\"volume\":\"57 3\",\"pages\":\"\"},\"PeriodicalIF\":2.7000,\"publicationDate\":\"2024-05-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Experimental Astronomy\",\"FirstCategoryId\":\"101\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s10686-024-09940-0\",\"RegionNum\":3,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ASTRONOMY & ASTROPHYSICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Experimental Astronomy","FirstCategoryId":"101","ListUrlMain":"https://link.springer.com/article/10.1007/s10686-024-09940-0","RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
DSRL: A low-resolution stellar spectral of LAMOST automatic classification method based on discrete wavelet transform and deep learning methods
Automatic classification of stellar spectra contributes to the study of the structure and evolution of the Milky Way and star formation. Currently available methods exhibit unsatisfactory spectral classification accuracy. This study investigates a method called DSRL, which is primarily used for automated and accurate classification of LAMOST stellar spectra based on MK classification criteria. The method utilizes discrete wavelet transform to decompose the spectra into high-frequency and low-frequency information, and combines residual networks and long short-term memory networks to extract both high-frequency and low-frequency features. By introducing self-distillation (DSRL-1, DSRL-2, and DSRL-3), the classification accuracy is improved. DSRL-3 demonstrates superior performance across multiple metrics compared to existing methods. In both three-class(F ,G ,K) and ten-class(A0, A5, F0, F5, G0, G5, K0, K5, M0, M5) experiments, DSRL-3 achieves impressive accuracy, precision, recall, and F1-Score results. Specifically, the accuracy performance reaches 94.50% and 97.25%, precision performance reaches 94.52% and 97.29%, recall performance reaches 94.52% and 97.22%, and F1-Score performance reaches 94.52% and 97.23%. The results indicate the significant practical value of DSRL in the classification of LAMOST stellar spectra. To validate the model, we visualize it using randomly selected stellar spectral data. The results demonstrate its excellent application potential in stellar spectral classification.
期刊介绍:
Many new instruments for observing astronomical objects at a variety of wavelengths have been and are continually being developed. Furthermore, a vast amount of effort is being put into the development of new techniques for data analysis in order to cope with great streams of data collected by these instruments.
Experimental Astronomy acts as a medium for the publication of papers of contemporary scientific interest on astrophysical instrumentation and methods necessary for the conduct of astronomy at all wavelength fields.
Experimental Astronomy publishes full-length articles, research letters and reviews on developments in detection techniques, instruments, and data analysis and image processing techniques. Occasional special issues are published, giving an in-depth presentation of the instrumentation and/or analysis connected with specific projects, such as satellite experiments or ground-based telescopes, or of specialized techniques.