FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2603671965> ?p ?o ?g. }

• W2603671965 abstract "Coronal Mass Ejections (CMEs) impact heavily on coronal activity,space weather and many interplanetary disturbance, so the detectionof CMEs are important for space weather disaster prevention and reduction.The traditional methods use man-made features or predefined thresholdto solve this problem. Despite the great progress in the detectionof CMEs, it is still a challenging problem due to the following threeaspects: Firstly, the early and late stage of the CMEs phenomenonis very weak, and the traditional image processing based method cannot detect this weak CMEs well. Secondly, the noises from comet, planetsand other stars can affect the detection of CMEs. Thirdly, the CMEsare complex and amorphous, and they are different in shapes, textures,grayscales, scales and so on. Because of these difficulties, it isdifficult to detect CMEs well by the traditional image processingmethod without modeling the CMEs. With the development of convolutionalneural networks (CNNs), it is possible to develop deep neural networksbased CMEs detection models to better solve this problem. For realizingthis, this paper presents an end-to-end detection method of CoronalMass Ejections detection: We design a deep neural network with 4 convolutionlayers, 1 full connection layer and 1 output layer. This deep neuralnetwork can automatically extract the image features that are suitableto describe the Coronal Mass Ejections, and can establish the CMEsdetection model based on the extracted features. In order to achievegood performance, we construct two datasets, one is mainly made upof strong CMEs, and the other is made up of weak or dark CMEs. Trainingis first done on the strong CMEs to obtain the initial CMEs detectionmodel. Based on the initial model established on the strong CMEs,finetuning is used on the weak CMEs to acquire the final CMEs detectionmodel. By using this scheme, training efficiency and good performancecan be guaranteed. In addition, the process is able to achieve selectionof features and setting of classification rules, which can realizeconveniently the end-to-end detection from data to results. The experimentalresults show that this method can effectively detect CMEs. The accuracyof our method is 100% on the strong CME dataset, and 91.54% on theweak CME dataset. The overall accuracy on our dataset is 98.05%. Finally,we test our method with the real data on May 2007 using LASCO catalogas the groudtruth. Experimental results show that our method achievesstate-of-the art performance comparing with the usually used catalog,SEEDS, CACTUS. The significance of our work is two-fold. On one hand,our method is end-to-end method which can select optimal featuresfor the detection task. According to the theory of deep neural networks,the selected features are usually those that the human vision uses.Therefore, the obtained features can depicts CMEs well and can beused by other methods. On the other hand, the good performance ofour method proves the depiction ability of the CNN for CME. So itis natural that CNN based description for the CME can also have goodperformance for the study of the modeling of the evolution of theCMEs." @default.
• W2603671965 created "2017-04-07" @default.
• W2603671965 creator A5003484876 @default.
• W2603671965 creator A5010848918 @default.
• W2603671965 creator A5058268034 @default.
• W2603671965 creator A5075965575 @default.
• W2603671965 creator A5089607926 @default.
• W2603671965 date "2017-03-02" @default.
• W2603671965 modified "2023-09-27" @default.
• W2603671965 title "An end-to-end method of Coronal Mass Ejections detection" @default.
• W2603671965 doi "https://doi.org/10.1360/n972016-00382" @default.
• W2603671965 hasPublicationYear "2017" @default.
• W2603671965 type Work @default.
• W2603671965 sameAs 2603671965 @default.
• W2603671965 citedByCount "0" @default.
• W2603671965 crossrefType "journal-article" @default.
• W2603671965 hasAuthorship W2603671965A5003484876 @default.
• W2603671965 hasAuthorship W2603671965A5010848918 @default.
• W2603671965 hasAuthorship W2603671965A5058268034 @default.
• W2603671965 hasAuthorship W2603671965A5075965575 @default.
• W2603671965 hasAuthorship W2603671965A5089607926 @default.
• W2603671965 hasBestOaLocation W26036719651 @default.
• W2603671965 hasConcept C108411613 @default.
• W2603671965 hasConcept C108583219 @default.
• W2603671965 hasConcept C115260700 @default.
• W2603671965 hasConcept C121332964 @default.
• W2603671965 hasConcept C1276947 @default.
• W2603671965 hasConcept C151325931 @default.
• W2603671965 hasConcept C154945302 @default.
• W2603671965 hasConcept C41008148 @default.
• W2603671965 hasConcept C50644808 @default.
• W2603671965 hasConcept C62520636 @default.
• W2603671965 hasConcept C72886185 @default.
• W2603671965 hasConcept C81363708 @default.
• W2603671965 hasConceptScore W2603671965C108411613 @default.
• W2603671965 hasConceptScore W2603671965C108583219 @default.
• W2603671965 hasConceptScore W2603671965C115260700 @default.
• W2603671965 hasConceptScore W2603671965C121332964 @default.
• W2603671965 hasConceptScore W2603671965C1276947 @default.
• W2603671965 hasConceptScore W2603671965C151325931 @default.
• W2603671965 hasConceptScore W2603671965C154945302 @default.
• W2603671965 hasConceptScore W2603671965C41008148 @default.
• W2603671965 hasConceptScore W2603671965C50644808 @default.
• W2603671965 hasConceptScore W2603671965C62520636 @default.
• W2603671965 hasConceptScore W2603671965C72886185 @default.
• W2603671965 hasConceptScore W2603671965C81363708 @default.
• W2603671965 hasLocation W26036719651 @default.
• W2603671965 hasOpenAccess W2603671965 @default.
• W2603671965 hasPrimaryLocation W26036719651 @default.
• W2603671965 hasRelatedWork W1480993447 @default.
• W2603671965 hasRelatedWork W2083660464 @default.
• W2603671965 hasRelatedWork W2139145365 @default.
• W2603671965 hasRelatedWork W2143542959 @default.
• W2603671965 hasRelatedWork W2154469849 @default.
• W2603671965 hasRelatedWork W2749694853 @default.
• W2603671965 hasRelatedWork W2762853287 @default.
• W2603671965 hasRelatedWork W2891029962 @default.
• W2603671965 hasRelatedWork W2950312287 @default.
• W2603671965 hasRelatedWork W2986640706 @default.
• W2603671965 hasRelatedWork W3004960979 @default.
• W2603671965 hasRelatedWork W3101881004 @default.
• W2603671965 hasRelatedWork W3106300692 @default.
• W2603671965 hasRelatedWork W3131017962 @default.
• W2603671965 hasRelatedWork W3165030534 @default.
• W2603671965 hasRelatedWork W3165290193 @default.
• W2603671965 hasRelatedWork W3175161600 @default.
• W2603671965 hasRelatedWork W3186460888 @default.
• W2603671965 hasRelatedWork W3195446334 @default.
• W2603671965 hasRelatedWork W3204959587 @default.
• W2603671965 isParatext "false" @default.
• W2603671965 isRetracted "false" @default.
• W2603671965 magId "2603671965" @default.
• W2603671965 workType "article" @default.