{"title":"将变异自动编码器应用于全磁盘太阳图像的无监督异常检测","authors":"Marius Giger, André Csillaghy","doi":"10.1029/2023sw003516","DOIUrl":null,"url":null,"abstract":"Deep learning is successful in many fields due to its ability to learn strong feature representations without the need for hand-crafted features, resulting in models with high representational power. However, many of these models are based on supervised learning and therefore depend on the availability of large annotated data sets. These are often difficult to obtain because they require human input. A common challenge for researchers in space weather is the sparsity of annotations in many of the available data sets, which are either unlabeled or have ambiguous labels. To alleviate the data bottleneck of loosely annotated data sets, unsupervised deep learning has become an important strategy, with anomaly detection being one of the most prominent applications. Unsupervised models have been successfully applied in various domains, such as medical imaging or video surveillance, to distinguish normal from abnormal data. In this work, we investigate how a purely unsupervised approach can be used to detect and extract solar phenomena in extreme ultraviolet images from the NASA SDO spacecraft. We show how a model based on variational autoencoders can be used to detect out-of-distribution samples and to localize regions of interest for solar activity. By using an unsupervised approach, we hope to contribute to space weather monitoring tools and further improve the understanding of space weather drivers.","PeriodicalId":22181,"journal":{"name":"Space Weather","volume":"21 1","pages":""},"PeriodicalIF":3.7000,"publicationDate":"2024-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Unsupervised Anomaly Detection With Variational Autoencoders Applied to Full-Disk Solar Images\",\"authors\":\"Marius Giger, André Csillaghy\",\"doi\":\"10.1029/2023sw003516\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Deep learning is successful in many fields due to its ability to learn strong feature representations without the need for hand-crafted features, resulting in models with high representational power. However, many of these models are based on supervised learning and therefore depend on the availability of large annotated data sets. These are often difficult to obtain because they require human input. A common challenge for researchers in space weather is the sparsity of annotations in many of the available data sets, which are either unlabeled or have ambiguous labels. To alleviate the data bottleneck of loosely annotated data sets, unsupervised deep learning has become an important strategy, with anomaly detection being one of the most prominent applications. Unsupervised models have been successfully applied in various domains, such as medical imaging or video surveillance, to distinguish normal from abnormal data. In this work, we investigate how a purely unsupervised approach can be used to detect and extract solar phenomena in extreme ultraviolet images from the NASA SDO spacecraft. We show how a model based on variational autoencoders can be used to detect out-of-distribution samples and to localize regions of interest for solar activity. By using an unsupervised approach, we hope to contribute to space weather monitoring tools and further improve the understanding of space weather drivers.\",\"PeriodicalId\":22181,\"journal\":{\"name\":\"Space Weather\",\"volume\":\"21 1\",\"pages\":\"\"},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2024-02-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Space Weather\",\"FirstCategoryId\":\"89\",\"ListUrlMain\":\"https://doi.org/10.1029/2023sw003516\",\"RegionNum\":2,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Space Weather","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1029/2023sw003516","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
深度学习之所以能在许多领域取得成功,是因为它能够学习强大的特征表征,而无需手工创建特征,从而产生具有高表征能力的模型。然而,这些模型中有许多是基于监督学习的,因此依赖于大量注释数据集的可用性。这些数据集通常很难获得,因为它们需要人工输入。空间天气研究人员面临的一个共同挑战是许多可用数据集的注释稀少,这些数据集要么没有标签,要么标签含糊不清。为了缓解松散注释数据集的数据瓶颈,无监督深度学习已成为一种重要策略,异常检测就是其中最突出的应用之一。无监督模型已成功应用于医疗成像或视频监控等多个领域,用于区分正常与异常数据。在这项工作中,我们研究了如何利用纯粹的无监督方法来检测和提取 NASA SDO 航天器极紫外图像中的太阳现象。我们展示了如何利用基于变异自动编码器的模型来检测异常分布样本,并定位太阳活动的相关区域。通过使用无监督方法,我们希望为空间天气监测工具做出贡献,并进一步提高对空间天气驱动因素的理解。
Unsupervised Anomaly Detection With Variational Autoencoders Applied to Full-Disk Solar Images
Deep learning is successful in many fields due to its ability to learn strong feature representations without the need for hand-crafted features, resulting in models with high representational power. However, many of these models are based on supervised learning and therefore depend on the availability of large annotated data sets. These are often difficult to obtain because they require human input. A common challenge for researchers in space weather is the sparsity of annotations in many of the available data sets, which are either unlabeled or have ambiguous labels. To alleviate the data bottleneck of loosely annotated data sets, unsupervised deep learning has become an important strategy, with anomaly detection being one of the most prominent applications. Unsupervised models have been successfully applied in various domains, such as medical imaging or video surveillance, to distinguish normal from abnormal data. In this work, we investigate how a purely unsupervised approach can be used to detect and extract solar phenomena in extreme ultraviolet images from the NASA SDO spacecraft. We show how a model based on variational autoencoders can be used to detect out-of-distribution samples and to localize regions of interest for solar activity. By using an unsupervised approach, we hope to contribute to space weather monitoring tools and further improve the understanding of space weather drivers.