基于SDO/HMI观测图像的相对太阳黑子数自动计算方法研究

doi:10.11871/jfdc.issn.2096-742X.2026.02.008

数据与计算发展前沿 ›› 2026, Vol. 8 ›› Issue (2): 98-110.

CSTR: 32002.14.jfdc.CN10-1649/TP.2026.02.008

doi: 10.11871/jfdc.issn.2096-742X.2026.02.008

基于SDO/HMI观测图像的相对太阳黑子数自动计算方法研究

赵星凯^1,²(),邹自明^1,^3,^*(),李云龙^1,³

¹ 中国科学院国家空间科学中心，北京 100190
² 中国科学院大学，北京 100049
³ 国家空间科学数据中心，北京 100190

收稿日期:2025-04-19 出版日期:2026-04-20 发布日期:2026-04-23
通讯作者: *邹自明（E-mail: mzou@nssc.ac.cn）
作者简介:赵星凯，中国科学院国家空间科学中心，硕士研究生，研究方向为太阳黑子的识别。
本文中负责论文撰写与实验论证。
ZHAO Xingkai is a master's student at the National Space Science Center, Chinese Academy of Sciences. His research interests include sunspot identification.
In this paper, he is responsible for Paper writing and experimental validation.
E-mail: zhaoxingkai22@mails.ucas.ac.cn|邹自明，中国科学院国家空间科学中心，博士，研究员，博士生导师，国家空间科学数据中心主任。主要研究方向为空间科学与数据科学交叉领域研究。
本文中负责论文架构设计，提出修改意见。
ZOU Ziming, Ph.D., is a Professor and Ph.D. supervisor at the National Space Science Center, CAS, and director of the National Space Science Data Center. His main research interests include the intersection of space science and data science.
In this paper, he is responsible for designing the structure of the article and making suggestions for revision.
E-mail: mzou@nssc.ac.cn

Research on the Automatic Calculation Method of Relative Sunspot Number Based on SDO/HMI White Light Images

ZHAO Xingkai^1,²(),ZOU Ziming^1,^3,^*(),LI Yunlong^1,³

¹ National Space Science Center, Chinese Academy of Sciences, Beijing 100190
² University of Chinese Academy of Sciences, Beijing 100049
³ National Space Science Data Center, Beijing 100190

Received:2025-04-19 Online:2026-04-20 Published:2026-04-23

摘要/Abstract

摘要：

【目的】 利用太阳动力学天文台（SDO）所拍摄的高分辨率白光图，自动计算相对太阳黑子数。【方法】 首先使用双重灰度阈值的分割算法，用于实现太阳黑子本影-半影结构的精确分割，以计算太阳黑子数Ns；随后采用基于密度的DBSCAN聚类算法对分割得到的黑子进行聚类，得到太阳黑子群数Ns；然后基于前述步骤获取的Ns和Ng，通过最小二乘法将计算结果与太阳影响数据分析中心发布的相对太阳黑子数进行拟合，确定出最优个人观测系数k；最后代入公式得到相对太阳黑子数R。【结果】 2014年与2017年的相对太阳黑子数计算结果与太阳影响数据分析中心数据的相关系数分别为0.913和0.960，RMSE分别为16.76和6.07。【局限】算法在不同太阳活动水平时期（高年与低年）表现出的差异导致个人观测系数k值波动较大，后续可能需要针对不同太阳活动水平设置动态的k值。【结论】 本文提出的算法实现了相对太阳黑子数的自动计算，极大改善了传统人工计算相对太阳黑子数依赖专家经验的问题，有望代替人工实现相对太阳黑子数的自动计算。

关键词: 太阳黑子, 相对太阳黑子数, 灰度阈值法, SDO/HMI

Abstract:

[Objective] To automatically calculate the relative sunspot number by using the high-resolution white light image captured by the Solar Dynamics Observatory (SDO). [Methods] Firstly, a segmentation algorithm with double grayscale thresholds is used to achieve the accurate segmentation of the umbra-penumbra structure of sunspots, so as to calculate the number of sunspots Ns. Subsequently, the density-based DBSCAN clustering algorithm is employed to cluster the segmented sunspots to obtain the number of sunspot groups Ng. Then, based on Ns and Ng obtained from the previous steps, the calculation results are fitted with the relative sunspot numbers released by the Solar Influences Data Analysis Center through the least squares method to determine the optimal personal observation coefficient k. Finally, the values are substituted into the formula to obtain the relative sunspot number R. [Results] The correlation coefficients between the calculated results of relative sunspot numbers in 2014 and 2017 and the data from the Solar Influences Data Analysis Center are 0.913 and 0.960, respectively, with RMSE values of 16.76 and 6.07, respectively. [Limitations] The performance of the algorithm varies during different solar activity periods (solar maximum and solar minimum), leading to significant fluctuations in individual observation coefficient k-values. Consequently, dynamic k-values tailored to different solar activity levels may need to be implemented in the future. [Conclusions] The algorithm proposed in this paper realizes the automatic calculation of the relative sunspot number, which greatly alleviates the problem that the traditional manual calculation of the relative sunspot number relies on expert experience. It is expected to replace manual work and realize the automatic calculation of the relative sunspot number.

Key words: Sunspot, relative sunspot number, thresholding method, SDO/HMI

赵星凯, 邹自明, 李云龙. 基于SDO/HMI观测图像的相对太阳黑子数自动计算方法研究[J]. 数据与计算发展前沿, 2026, 8(2): 98-110.

ZHAO Xingkai, ZOU Ziming, LI Yunlong. Research on the Automatic Calculation Method of Relative Sunspot Number Based on SDO/HMI White Light Images[J]. Frontiers of Data and Computing, 2026, 8(2): 98-110, https://cstr.cn/32002.14.jfdc.CN10-1649/TP.2026.02.008.

图/表 14

图1

图2

图3

图4

图5

图6

图7

图8

表1

表2

图9

图10

图11

图12

参考文献 27

[1]	USOSKIN I. A history of solar activity over millennia[J]. LIVING REVIEWS IN SOLAR PHYSICS, 2017, 14(1): 3. DOI: 10.1007/s41116-017-0006-9.
[2]	CLETTE F, SVALGAARD L, VAQUERO J M, et al. Revisiting the Sunspot Number A 400-Year Perspective on the Solar Cycle[J]. Space Science Reviews, 2014, 186(1-4): 35-103. DOI:10.1007/s11214-014-0074-2.
[3]	VANLOMMEL P, CUGNON P, LINDEN R A M V D, et al. The Sidc: World Data Center for the Sunspot Index[J]. Solar Physics, 2004, 224(1-2): 113-120. DOI: 10.1007/s11207-005-6504-2.
[4]	CLETTE F, LEFÈVRE L. The New Sunspot Number: Assembling All Corrections[J]. Solar Physics, 2016, 291(9-10): 2629-2651. DOI:10.1007/s11207-016-1014-y.
[5]	VAQUERO J M. Historical sunspot observations: A review[J]. Advances in Space Research, 2007, 40(7): 929-941. DOI:10.1016/j.asr.2007.01.087.
[6]	CORTESI S, CAGNOTTI M, BIANDA M, et al. Sunspot Observations and Counting at Specola Solare Ticinese in Locarno Since 1957[J]. Solar Physics, 2016, 291(9-10):3075-3080.DOI:10.1007/s11207-016-0872-7.
[7]	ZHARKOV S, ZHARKOVA V, IPSON S, et al. Technique for automated recognition of sunspots on full-disk solar images[J]. EURASIP Journal on Advances in Signal Processing, 2005, 15: 2573-2584. DOI: 10.1155/ASP.2005.2573.
[8]	CURTO J J, BLANCA M, MARTÍNEZ E. Automatic Sunspots Detection on Full-Disk Solar Images using Mathematical Morphology[J]. Solar Physics, 2008, 250(2): 411-429. DOI: 10.1007/s11207-008-9224-6.
[9]	ZHAO C, LIN G, DENG Y, et al. Automatic Recognition of Sunspots in HSOS Full-Disk Solar Images[J]. Publications of the Astronomical Society of Australia, 2016, 33: e018. DOI: 10.1017/pasa.2016.17.
[10]	COLAK T, QAHWAJI R. Automated McIntosh-Based classification of sunspot groups using MDI images[J]. Solar Physics, 2008, 248(2): 277-296. DOI:10.1007/s11207-007-9094-3.
[11]	李泠, 崔延美, 刘四清, 等. 太阳黑子自动识别与特征参量自动提取[J]. 空间科学学报, 2020, 40(3): 315-322.
[12]	ZHAO C, YANG S, WANG T, et al. An Automatic Approach for Grouping Sunspots and Calculating Relative Sunspot Number on SDO/HMI Continuum Images[J]. The Astronomical Journal, 2024, 167(2): 52. DOI: 10.3847/1538-3881/ad11e2.
[13]	BOURGEOIS S, BARATA T, ERDÉLYI R, et al. Sunspots Identification Through Mathematical Morphology[J]. Solar Physics, 2024, 299(2): 10. DOI: 10.1007/s11207-023-02243-1.
[14]	MOURATO A, FARIA J, VENTURA R. Automatic sunspot detection through semantic and instance segmentation approaches[J]. Engineering Applications of Artificial Intelligence, 2024, 129: 107636. DOI:10.1016/j.engappai.2023.107636.
[15]	FAN D, YANG Y, FENG S, et al. SIPNet & SAHI: Multiscale Sunspot Extraction for High-Resolution Full Solar Images[J]. Applied Sciences, 2023, 14(1): 7. DOI: 10.3390/app14010007.
[16]	SANTOS J, PEIXINHO N, BARATA T, et al. Sunspot Detection Using YOLOv5 in Spectroheliograph H-Alpha Images[J]. Applied Sciences, 2023, 13(10): 5833. DOI: 10.3390/app13105833.
[17]	赵梓良, 刘家真, 胡真, 等. 一种层次化的太阳黑子快速自动识别方法[J]. 光电工程, 2020, 47(7): 39-49.
[18]	PALLADINO L, NTAGIOU E, KLUG J, et al. Sunspot Groups Detection and Classification on SDO/HMI Images using Deep Learning Techniques[C]// 2022 IEEE Aerospace Conference (AERO). New York: IEEE, 2022: 1-10. https://ieeexplore.ieee.org/document/9843222/. DOI: 10.1109/AERO53065.2022.9843222.
[19]	FINI. A deep learning approach to sunspot detection and counting[D]. Politecnico di Milano, 2019. https://www.politesi.polimi.it/handle/10589/147435.
[20]	DASGUPTA U, SINGH S, JEWALIKAR V. Sunspot Number Calculation Using Clustering[C]// 2011 Third National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, 2011: 171-174. https://ieeexplore.ieee.org/document/6133028/. DOI: 10.1109/NCVPRIPG.2011.43.
[21]	Which time tag should be attached to each daily International Sunspot Number? \| SIDC[EB]. https://www.sidc.be/SILSO/faq8.
[22]	LIN G H, WANG X F, LIU S, et al. Chinese Sunspot Drawings and Their Digitization -(I) Parameter Archives[J]. Solar Physics, 2019, 294(6): 79. DOI: 10.1007/s11207-019-1456-0.
[23]	BARANYI T, GYORI L, LUDMANY A, et al. Comparison of sunspot area data bases[J]. Monthly Notices of the Royal Astronomical Society, 2001, 323(1): 223-230. DOI: 10.1046/j.1365-8711.2001.04195.x.
[24]	BERGHMANS D, CLETTE F, CUGNON P, et al. The solar influences data analysis centre[J]. Journal of Atmospheric and Solar-Terrestrial Physics, 2002, 64(5-6): 757-761. DOI: 10.1016/S1364-6826(02)00037-8.
[25]	AYDIN B, ANGRYK R. Spatiotemporal Frequent Pattern Mining on Solar Data:Current Algorithms and Future Directions[C]// 2015 IEEE International Conference on Data Mining Workshop (ICDMW). Atlantic City, NJ, USA: IEEE, 2015: 575-581. http://ieeexplore.ieee.org/document/7395719/. DOI: 10.1109/ICDMW.2015.10.
[26]	OTSU N. Threshold Selection Method from Gray-Level Histograms[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9(1): 62-66. DOI: 10.1109/TSMC.1979.4310076.
[27]	DUAN L, XU L, GUO F, et al. A local-density based spatial clustering algorithm with noise[J]. Information Systems, 2007, 32(7): 978-986. DOI: 10.1016/j.is.2006.10.006.

Eps	70	80	90
MinPts
15	3.36	2.89	2.59
20	2.65	2.3	2.05
25	2.18	1.92	1.72

Eps	85	95	100
MinPts
25	1.82	1.69	1.65
30	1.61	1.56	1.56
35	1.56	1.55	1.58

基于SDO/HMI观测图像的相对太阳黑子数自动计算方法研究

Research on the Automatic Calculation Method of Relative Sunspot Number Based on SDO/HMI White Light Images

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 27

相关文章 0

编辑推荐

Metrics

本文评价