基于EfficientNet的星系形态分类研究

星系的结构和形态能够反映星系自身的物理性质,其形态的分类是后续分析研究的一个重要环节.EfficientNet模型使用复合系数对深度网络模型的深度、宽度、输入图像分辨率进行更加结构化的统一缩放,是一种新的深度网络优化扩展方法.将该模型应用于星系数据形态的分类研究中,结果表明基于EfficientNetB5模型的平均准确率、精确率、召回率以及F1分数(精确率与召回率的调和平均数)都在96.6%以上,与残差网络(Residual network, ResNet)中ResNet-26模型的分类结果相比有较大的提升.实验结果证明EfficientNet的深度网络优化扩展方法可行且有效,可应用于星系的形态分类.     

基于深度残差网络的星系形态分类

星系形态与星系的形成和演化有着密切的联系,因此星系形态分类(galaxy morphology classification)成为研究不同星系物理特征的重要过程之一。斯隆数字巡天(Sloan Digital Sky Survey, SDSS)等大型巡天计划产生的海量星系图像数据对星系形态的准确、实时分类提出了新的挑战,而深度学习(deep learning)算法能有效应对这类海量星系图片的自动分类考验。面向星系形态分类问题提出了一种改进的深度残差网络(residual network, ResNet),即ResNet-26模型。该模型对残差单元进行改进,减少了网络深度,并增加了网络宽度,实现了对星系形态特征的自动提取、识别和分类。实验结果表明,与Dieleman和ResNet-50等其他流行的卷积神经网络(convolution neural network, CNN)模型相比,ResNet-26模型具有更优的分类性能,可应用于未来大型巡天计划的大规模星系形态分类系统。     

EfficientNetV2-S-Triplet7: 一种改进的星系newline 形态学分类算法

星系的形态与星系的形成和演化息息相关, 其形态学分类是星系天文学后续研究的重要一环. 当前海量天文观测数据的出现使得天文数据自动分析方法越来越得到重视, 针对此问题, 利用先进的深度学习骨干网络EfficientNetV2, 分析不同的注意力机制类型和使用节点对网络性能的影响, 构建了一种命名为EfficientNetV2-S-Triplet7 (即在EfficientNetV2-S stage7的$1\times1$卷积层后加入Triplet模块)的改进算法模型来实现星系形态学的自动分类. 使用第二期星系动物园(Galaxy Zoo 2, GZ2)中超过24万张的测光图像作为初始数据进行实验测试. 在对数据进行预处理时采取了尺寸抖动、翻转、色彩畸变等图像增强手段来解决图像数量的不平衡问题. 在同一系列经典和前沿的深度学习算法模型AlexNet、ResNet-34、MobileNetV2、RegNet进行对比实验后, 得出EfficientNetV2-S-Triplet7算法在分类准确率、查全率和F1分数等指标上具有最好的测试结果. 在9375张测试图像中的3项指标值分别可达到89.03%、90.21%、89.93%, 查准率达到89.69%, 在其他模型中排在第3位. 该结果表明将EfficientNetV2-S-Triplet7算法应用于大规模星系数据的形态学分类任务中有很好的效果.     

基于Stacking集成学习的恒星/星系分类研究

机器学习在当今诸多领域已经取得了巨大的成功,但是机器学习的预测效果往往依赖于具体问题.集成学习通过综合多个基分类器来预测结果,因此,其适应各种场景的能力较强,分类准确率较高.基于斯隆数字巡天(Sloan Digital Sky Survey,SDSS)计划恒星/星系中最暗源星等集分类正确率低的问题,提出一种基于Stacking集成学习的恒星/星系分类算法.从SDSS-DR7(SDSS Data Release 7)中获取完整的测光数据集,并根据星等值划分为亮源星等集、暗源星等集和最暗源星等集.仅针对分类较为复杂且困难的最暗源星等集展开分类研究.首先,对最暗源星等集使用10折嵌套交叉验证,然后使用支持向量机(Support Vector Machine,SVM)、随机森林(Random Forest,RF)、XGBoost(eXtreme Gradient Boosting)等算法建立基分类器模型;使用梯度提升树(Gradient Boosting Decision Tree,GBDT)作为元分类器模型.最后,使用基于星系的分类正确率等指标,与功能树(Function Tree,FT)、SVM、RF、GBDT、XGBoost、堆叠降噪自编码(Stacked Denoising AutoEncoders,SDAE)、深度置信网络(Deep Belief Network,DBN)、深度感知决策树(Deep Perception Decision Tree,DPDT)等模型进行分类结果对比分析.实验结果表明,Stacking集成学习模型在最暗源星等集分类中要比FT算法的星系分类正确率提高了将近10%.同其他传统的机器学习算法、较强的提升算法、深度学习算法相比,Stacking集成学习模型也有较大的提升.     

一种低表面亮度星系的自动搜索算法newline--YOLOX-CS

低表面亮度星系(Low Surface Brightness Galaxy, LSBG)的特征对于理解星系整体特征非常重要, 通过现代的机器学习特别是深度学习算法来搜寻扩充低表面亮度星系样本具有重要意义. LSBG因特征不明显而难以用传统方法进行自动和准确辨别, 但深度学习确具有自动找出复杂且有效特征的优势, 针对此问题提出了一种可用于在大样本巡天观测项目中搜寻LSBG的算法---YOLOX-CS (You Only Look Once version X-CS). 首先通过实验对比5种经典目标检测算法并选择较优的YOLOX算法作为基础算法, 然后结合不同注意力机制和不同优化器, 构建了YOLOX-CS的框架结构. 数据集使用的是斯隆数字化巡天(Sloan Digital Sky Survey, SDSS)中的图像, 其标签来自于$\alpha.40$-SDSS DR7 (40%中性氢苜蓿巡天与第7次数据发布的斯隆数字化巡天的交叉覆盖天区)巡天项目中的LSBG, 由于该数据集样本较少, 还采用了深度卷积生成对抗网络(Deep Convolutional Generative Adversarial Networks, DCGAN)模型扩充了实验测试数据. 通过与一系列目标检测算法对比后, YOLOX-CS在扩充前后两个数据集中搜索LSBG的召回率和AP (Average Precision)值都有较好的测试结果, 其在未扩充数据集的测试集中的召回率达到97.75%, AP值达到97.83%, 在DCGAN模型扩充的数据集中, 同样测试集下进行实验的召回率达到99.10%, AP值达到98.94%, 验证了该算法在LSBG搜索中具有优秀的性能. 最后, 将该算法应用到SDSS部分测光数据上, 搜寻得到了765个LSBG候选体.     

基于卷积神经网络的恒星光谱型和光度型的分类模型

恒星光谱分类是天文学中一个重要的研究问题.对于已经采集到的海量高维恒星光谱数据的分类,采用模式匹配方法对光谱型分类较为成功,但其缺点在于标准恒星模版之间的差异性在匹配实际观测数据中不能体现出来,尤其是当需要进行光谱型和光度型的二元分类时模版匹配法往往会失败.而采用谱线特征测量的光度型分类强烈地依赖谱线拟合的准确性.为了解决二元分类的问题,介绍了一种基于卷积神经网络的恒星光谱型和光度型分类模型(Classification model of Stellar Spectral type and Luminosity type based on Convolution Neural Network, CSSL CNN).这一模型使用卷积神经网络来提取光谱的特征,通过注意力模块学习到了重要的光谱特征,借助池化操作降低了光谱的维度并压缩了模型参数的数量,使用全连接层来学习特征并对恒星光谱进行分类.实验中使用了大天区面积多目标光纤光谱天文望远镜(Large Sky Area Multi-Object Fiber Spectroscopy Telescope, LAMOST)公开数据集Data Release 5 (DR5,用了其中71282条恒星光谱数据,每条光谱包含了3000多维的特征)对该模型的性能进行验证与评估.实验结果表明,基于卷积神经网络的模型在恒星的光谱型分类上准确率达到92.04%,而基于深度神经网络的模型(Celestial bodies Spectral Classification Model, CSC Model)只有87.54%的准确率; CSSL CNN在恒星的光谱型和光度型二元分类上准确率达到83.91%,而模式匹配方法MKCLASS仅有38.38%的准确率且效率较低.     

基于NPSVM模型的超新星识别方法

巡天观测与高能物理、黑洞天文等领域均有密切的联系.基于星系-超新星二分类问题,研究光谱数据预处理,结合余弦相似度改善PCA(Principal Component Analysis)光谱分解特征提取方法,用SDSS(the Sloan Digital Sky Survey)、WISeREP(the Weizmann Interactive Supernova data REPository)组成的5620条光谱数据集训练支持向量机,可以得到0.498%泛化误差的识别模型和新样本分类概率.使用Neyman-Pearson决策方法建立NPSVM(Neyman-Pearson Support Vector Machine)模型可进一步降低超新星的漏判率.     

基于胶囊网络的恒星光谱分类研究

大型巡天项目的快速发展,产生大量的恒星光谱数据,也使得实现恒星光谱数据的自动分类成为一项具有挑战性的工作.提出一种新的基于胶囊网络的恒星光谱分类方法,首先利用1维卷积网络和短时傅里叶变换将来源于LAMOST(Large Sky Area Multi-Object Fiber Spectroscopy Telescope)Data Release 5(DR5)的F5、G5、K5型1维恒星光谱转化成2维傅里叶谱图像,再通过胶囊网络对2维谱图像进行自动分类.由于胶囊网络具有保留图像中实体之间的分层位姿关系和无需池化层的优点,实验结果表明:胶囊网络具有较好的分类性能,对于F5、G5、K5型恒星光谱的分类,准确率优于其他分类方法.     

早型星系的恒星形成活动演化研究

从观测上测定早型星系中恒星形成活动随红移的演化有助于理解这类星系的形成演化.结合GEMS（Galaxy Evolution from Morphology and SEDs）巡天的HST/ACS（Hubble Space Telescope/Advanced Camera for Surveys）高分辨图像和CDFS（ChandraDeep Field South）天区Spitzer、GALEX（Galaxy Evolution Explorer）等多波段数据,基于形态、颜色和恒星质量选出一个0.2≤z≤1.0红移范围的包含456个早型星系的完备样本.利用stacking技术测量了样本星系紫外与红外平均光度,估计早型星系的恒星形成率.结果显示,早型星系中的恒星形成率较低(<3 M_⊙·yr^-1),随红移递减而降低.在红移z=1以来的恒星形成贡献的质量小于15%.星族分析亦肯定大质量早型星系的主体星族形成于宇宙早期（z>2）.     

基于XGBoost算法的恒星/星系分类研究

机器学习在当今的诸多领域已经取得了巨大的成功.尤其是提升算法.提升算法适应各种场景的能力较强、准确率较高,已经在多个领域发挥巨大的作用.但是提升算法在天文学中的应用却极为少见.为解决斯隆数字巡天(Sloan Digital Sky Survey,SDSS)数据中恒星/星系暗源集分类正确率低的问题,引入了机器学习中较新的研究成果–XGBoost (eXtreme Gradient Boosting).从SDSS-DR7 (SDSS Data Release 7)中获取完整的测光数据集,并根据星等值划分为亮源集和暗源集.首先,分别对亮源集和暗源集使用十折交叉验证法,同时运用XGBoost算法建立恒星/星系分类模型;然后,运用栅格搜索等方法调优XGBoost参数;最后,基于星系的分类正确率等指标,与功能树(Function Tree, FT)、Adaboost (Adaptive boosting)、随机森林(Random Forest, RF)、梯度提升决策树(Gradient Boosting Decision Tree, GBDT)、堆叠降噪自编码(Stacked Denoising AutoEncoders, SDAE)、深度置信网络(Deep Belief Network, DBN)等模型进行对比并分析结果.实验结果表明:XGBoost在暗源分类中要比功能树算法的星系分类正确率提高了将近10%,在暗源集的最暗星等中比功能树提高了将近5%.同其他传统的机器学习算法和深度神经网络相比, XGBoost也有不同程度的提升.     

Study of Star/Galaxy Classification Based on the XGBoost Algorithm

Machine learning has achieved great success in many areas today. The lifting algorithm has a strong ability to adapt to various scenarios with a high accuracy, and has played a great role in many fields. But in astronomy, the application of lifting algorithms is still rare. In response to the low classification accuracy of the dark star/galaxy source set in the Sloan Digital Sky Survey (SDSS), a new research result of machine learning, eXtreme Gradient Boosting (XGBoost), has been introduced. The complete photometric data set is obtained from the SDSS-DR7, and divided into a bright source set and a dark source set according to the star magnitude. Firstly, the ten-fold cross-validation method is used for the bright source set and the dark source set respectively, and the XGBoost algorithm is used to establish the star/galaxy classification model. Then, the grid search and other methods are used to adjust the XGBoost parameters. Finally, based on the galaxy classification accuracy and other indicators, the classification results are analyzed, by comparing with the models of function tree (FT), Adaptive boosting (Adaboost), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), Stacked Denoising AutoEncoders (SDAE), and Deep Belief Nets (DBN). The experimental results show that, the XGBoost improves the classification accuracy of galaxies in the dark source classification by nearly 10% as compared to the function tree algorithm, and improves the classification accuracy of sources with the darkest magnitudes in the dark source set by nearly 5% as compared to the function tree algorithm. Compared with other traditional machine learning algorithms and deep neural networks, the XGBoost also has different degrees of improvement.     

Research on Star/Galaxy Classification Based on Stacking Ensemble Learning

Machine learning has achieved great success in many areas today, but the forecast effect of machine learning often depends on the specific problem. An ensemble learning forecasts results by combining multiple base classifiers. Therefore, its ability to adapt to various scenarios is strong, and the classification accuracy is high. In response to the low classification accuracy of the darkest source magnitude set of stars/galaxies in the Sloan Digital Sky Survey (SDSS), a star/galaxy classification algorithm based on the stacking ensemble learning is proposed in this paper. The complete photometric data set is obtained from the SDSS Data Release (DR) 7, and divided into the bright source magnitude set, dark source magnitude set, and darkest source magnitude set according to the stellar magnitude. Firstly, the 10-fold nested cross-validation method is used for the darkest source magnitude set, then the Support Vector Machine (SVM), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) algorithms are used to establish the base-classifier model; the Gradient Boosting Decision Tree (GBDT) is used as the meta-classifier model. Finally, based on the classification accuracy of galaxies and other indicators, the classification results are analyzed and compared with the results obtained by the Function Tree (FT), SVM, RF, GBDT, Stacked Denoising Autoencoders (SDAE), Deep Belief Nets (DBN), and Deep Perception Decision Tree (DPDT) models. The experimental results show that the stacking ensemble learning model has improved the classification accuracy of galaxies in the darkest source magnitude set by nearly 10% compared to the function tree algorithm. Compared with other traditional machine learning algorithm, stronger lifting algorithm, and deep learning algorithm, the stacking ensemble learning model also has different degrees of improvement.     

PSCz superclusters: detection, shapes and cosmological implications

We study the possibility of correctly identifying, from the smooth galaxy density field of the PSC z flux-limited catalogue, high-density regions (superclusters) and recovering their true shapes in the presence of a bias introduced by the coupling between the selection function and the constant radius smoothing. We quantify such systematic biases in the smoothed PSC z density field and after applying the necessary corrections we study supercluster multiplicity and morphologies using a differential geometry definition of shape. Our results strongly suggest that filamentary morphology is the dominant feature of PSC z superclusters. Finally, we compare our results with those expected in three different cosmological models and find that the Λ cold dark matter (CDM) model (Ω_Λ=1−Ω_m=0.7) performs better than Ω_m=1 CDM models.     

Morphology and Structure of High-redshift Massive Galaxies in the CANDELS Fields

Using the multi-band photometric data of all five CANDELS (Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey) fields and the near-infrared (F125W and F160W) high-resolution images of HST WFC3 (Hubble Space Telescope Wide Field Camera 3), a quantitative study of morphology and structure of mass-selected galaxies is presented. The sample includes 8002 galaxies with a redshift 1 < z < 3 and stellar mass M_*> 10¹⁰M_⊙. Based on the Convolutional Neural Network (ConvNet) criteria, we classify the sample galaxies into SPHeroids (SPH), Early-Type Disks (ETD), Late-Type Disks (LTD), and IRRegulars (IRR) in different redshift bins. The findings indicate that the galaxy morphology and structure evolve with redshift up to z ～ 3, from irregular galaxies in the high-redshift universe to the formation of the Hubble sequence dominated by disks and spheroids. For the same redshift interval, the median values of effective radii (r_e) of different morphological types are in a descending order: IRR, LTD, ETD, and SPH. But for the Sérsic index (n), the order is reversed (SPH, ETD, LTD, and IRR). In the meantime, the evolution of galaxy size (r_e) with the redshift is explored for the galaxies of different morphological types, and it is confirmed that their size will enlarge with time. However, such a phenomenon is not found in the relations between the redshift (1 < z < 3) and the mean axis ratio (b/a), as well as the Sérsic index (n).     

Objective classification of galaxy spectra using the information bottleneck method

A new method for classification of galaxy spectra is presented, based on a recently introduced information theoretical principle, the information bottleneck . For any desired number of classes, galaxies are classified such that the information content about the spectra is maximally preserved. The result is classes of galaxies with similar spectra, where the similarity is determined via a measure of information. We apply our method to ∼6000 galaxy spectra from the ongoing 2dF redshift survey, and a mock-2dF catalogue produced by a cold dark matter (CDM) based semi-analytic model of galaxy formation. We find a good match between the mean spectra of the classes found in the data and in the models. For the mock catalogue, we find that the classes produced by our algorithm form an intuitively sensible sequence in terms of physical properties such as colour, star formation activity, morphology, and internal velocity dispersion. We also show the correlation of the classes with the projections resulting from a principal component analysis.