恒星光谱数据弱特征识别方法Method for Recognizing Weak Features of Stellar Spectral Data
贺艳婷,周嘉炜,杨雨晴,贾凯雪,唐文龙,杨海峰
摘要(Abstract):
恒星光谱弱特征识别是LAMOST光谱数据分析的重要研究内容,能够为恒星光谱分类提供重要科学依据。目前,针对恒星光谱数据进行特征识别的方法较多,但是缺乏对某种特定特征谱线进行精确提取的算法。针对LAMOST低分辨光谱数据中Hα弱发射线轮廓形态多样问题,提出了一种基于置信度的Hα弱发射线识别方法。首先,基于Hα弱发射线轮廓形态特征给出Hα弱发射线的置信度的度量方法。利用Hα发射线波长区间内峰值与发射线的偏移量建立距离置信度模型,根据高斯轮廓所含像素点个数建立高斯轮廓副信息模型,通过计算峰值左右波形的差异建立对称性评估模型,结合三个模型给出最终的Hα弱发射线的置信度,并基于此置信度进行第一轮筛选。为了提高精度,提出了借助其它发射线的特征给出了基于二分类的Hα发射线筛选策略。通过考察Hβ、NII、OIII以及SII发射线的特征,基于辅助信息的决策树进行第二轮筛选,进一步提高筛选的精度。实验结果表明:提出的Hα弱发射线的特征度量方法的准确度高达90%,并且速度较快,平均每1k数据耗时仅三十多秒。
关键词(KeyWords): 决策树;二元分类;置信度;弱发射线;LAMOST光谱数据
基金项目(Foundation): 国家自然科学基金(U1931209);; 大学生创新训练项目(XJ2020092)
作者(Author): 贺艳婷,周嘉炜,杨雨晴,贾凯雪,唐文龙,杨海峰
参考文献(References):
- [1] BORNE K D.Next Generation of Data Mining[M].Boca Raton:CRC Press,2008:91-114.
- [2] LIU C,CUI W Y,ZHANG B,et al.Spectral classification of stars based on LAMOST spectra[J].Research in Astronomy and Astrophysics,2015,15(8):1137-1153.
- [3] NOLAN L A,HARVA M O,KABáN A,et al.A data-driven Bayesian approach for finding young stellar populations in early-type galaxies from their ultraviolet-optical spectra[J].Monthly Notices of the Royal Astronomical Society,2006,366(1):321-338.
- [4] YANG Y,CAI J,YANG H,et al.TAD:A trajectory clustering algorithm based on spatial-temporal density analysis[J].Expert Systems with Application,2020,139:112846.1-112846.16.
- [5] KIM E J,BRUNNER R J.Star-galaxy Classification Using Deep Convolutional Neural Networks[J].Monthly Notices of the Royal Astronomical Society,2017,464(4)4463-4475.
- [6] KOHOUTEK L,WEHMEYER R.Catalogue of H-alpha emission stars in the Northern Milky Way[J].Astronomy and Astrophysics-Berlin-,1999,342(3):910-910.
- [7] WITHAM A R,KNIGGE C,DREW J E,et al.The IPHAS catalogue of Hα emission-line sources in the northern Galactic plane[J].Monthly Notices of the Royal Astronomical Society,2010,384(4):1277-1288.
- [8] SIGUT T,PATEL P.The correlation between Hα emission and visual magnitude during long-term variations in classical be Stars[J].Astrophysical Journal,2013,765(1):930-940.