基于距离度量的LAMOST光谱中连续谱异常的自动检测

发布时间：2018-10-07 18:21

【摘要】：连续谱异常是指恒星光谱在获得和处理过程中由于星际消光和流量定标等原因造成连续谱严重偏离甚至中断的现象,这对光谱的谱线提取以及其他一些后续处理工作带来负面影响。提出了一种基于距离度量的连续谱异常光谱的自动检测方法,相比传统人眼检查在保证正确率的情况下大大地提高了工作效率。该方法首先通过光谱的lick线指数来确定待测光谱的恒星类型,同时对待测光谱进行归一化处理;然后分别提取待测光谱和对应类型模板光谱的连续谱;最后进行连续谱模板匹配,在每个波长点计算待测光谱和其模板光谱的流量差值,分析流量差值的分布,检验有多少差值点分布在在均值(β)附近的±α个标准差(δ)的范围内,进而可确定是否有连续谱异常。实验表明提出方法的可以快速有效的识别出连续谱异常的恒星光谱。
[Abstract]:Continuous spectrum anomaly refers to the serious deviation and even interruption of continuous spectrum caused by interstellar extinction and flow calibration during the acquisition and processing of stellar spectra. This has a negative effect on spectral line extraction and other subsequent processing. This paper presents an automatic detection method of abnormal spectrum of continuous spectrum based on distance measurement. Compared with traditional human eye inspection, it can improve the working efficiency greatly. In this method, the star type of the spectrum is determined by the lick line index of the spectrum, and the measured spectrum is normalized, then the continuous spectrum of the measured spectrum and the corresponding template spectrum are extracted respectively. Finally, the continuous spectral template matching is carried out, the flux difference between the measured spectrum and its template spectrum is calculated at each wavelength point, the distribution of the flow difference is analyzed, and the number of difference points is tested in the range of 卤伪 standard deviation (未) near the mean value (尾). Furthermore, it is possible to determine whether there is a continuous spectrum anomaly. The experimental results show that the proposed method can quickly and effectively identify the abnormal spectra of stars.
【作者单位】：山东大学(威海)机电与信息工程学院;中国科学院光学天文重点实验室国家天文台;
【基金】：国家自然科学基金项目(U1431102)资助
【分类号】：P111;P144.1

【相似文献】