Edge-Inferring Graph Neural Network with Dynamic Task-Guided Self-Diagnosis for Fewshot Hyperspectral Image Classification

Simple item page
dc.contributor.authorYu, Chunyan
dc.contributor.authorHuang, Jiahui
dc.contributor.authorSong, Meiping
dc.contributor.authorWang, Yulei
dc.contributor.authorChang, Chein-I
dc.date.accessioned2022-08-23T14:44:45Z
dc.date.available2022-08-23T14:44:45Z
dc.date.issued2022-08-04
dc.description.abstractThe current hyperspectral image classification (HSIC) model based on the convolutional neural network for feature extraction and softmax classifier has been prone to the barrier of label prediction with limited samples. Substituting for the enormously complicated work of terrain labeling, few-shot learning provides a popular option for HSIC with very few annotated samples. In this article, we proposed a novel edge-inferring framework with the metalearning paradigm for hyperspectral few-shot classification (HSFSC), in which a graph neural network for similarity measurement is first presented to iteratively infer edge labels with the exploitation of instance-level similarity and the distribution-level similarity. Besides, in the metatraining stage, the pixel prediction model and the patch prediction model based on edge-inferring architecture are concretized jointly to improve the classification accuracy of the test samples. Expressly, at the metatesting phase, the dynamic task-guided self-diagnosis strategy is developed for the first time to diagnose the samples separability of the current classification task, which is responsible for dynamically assigning the most reliable results based on the generated reliability grade of the sample. The extensive experimental results and analysis of three hyperspectral image datasets demonstrate the superiority of the proposed HSFSC architecture compared with other advanced methods.en_US
dc.description.sponsorshipThe work of Chunyan Yu was supported by the Science Foundation of Liaoning Province (Surface Project) under Grant LJKZ0065. The work of Meiping Song was supported in part by the National Nature Science Foundation of China under Grant 61971082 and in part by the Fundamental Research Funds for the Central Universities under Grant 3132017124.en_US
dc.description.urihttps://ieeexplore.ieee.org/abstract/document/9849702en_US
dc.format.extent13 pagesen_US
dc.genrejournal articlesen_US
dc.genrepostprintsen_US
dc.identifierdoi:10.13016/m2xnjn-h9xd
dc.identifier.citationC. Yu, J. Huang, M. Song, Y. Wang and C. -I. Chang, "Edge-Inferring Graph Neural Network with Dynamic Task-Guided Self-Diagnosis for Few-shot Hyperspectral Image Classification," in IEEE Transactions on Geoscience and Remote Sensing, 2022, doi: 10.1109/TGRS.2022.3196311.en_US
dc.identifier.urihttps://doi.org/10.1109/TGRS.2022.3196311
dc.identifier.urihttp://hdl.handle.net/11603/25547
dc.language.isoen_USen_US
dc.publisherIEEEen_US
dc.relation.isAvailableAtThe University of Maryland, Baltimore County (UMBC)
dc.relation.ispartofUMBC Computer Science and Electrical Engineering Department Collection
dc.relation.ispartofUMBC Faculty Collection
dc.rights© 2022 IEEE.  Personal use of this material is permitted.  Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.en_US
dc.titleEdge-Inferring Graph Neural Network with Dynamic Task-Guided Self-Diagnosis for Fewshot Hyperspectral Image Classificationen_US
dc.typeTexten_US
dcterms.creatorhttps://orcid.org/0000-0002-5450-4891en_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Edge-Inferring_Graph_Neural_Network_With_Dynamic_Task-Guided_Self-Diagnosis_for_Few-Shot_Hyperspectral_Image_Classification.pdf
Size:
3.67 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.56 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

UMBC Computer Science and Electrical Engineering Department
UMBC Faculty Collection