FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4361801444> ?p ?o ?g. }

• W4361801444 endingPage "14" @default.
• W4361801444 startingPage "1" @default.
• W4361801444 abstract "Deep learning approaches have been extensively applied to change detection in hyperspectral images (HSIs). However, the majority of them encounter scarcity of training samples or rely on complex structures and learning strategies. Although untrained change detection models have been proved to be effective in relief above problems, they were constructed using regular convolutions and treated spatial locations and channels equally, which are insufficient to extract discriminative features and lead to limited accuracy. Given this, a novel untrained framework using randomly initialized models with spatial-channel augmentation (RICD) is proposed for HSI change detection in this paper. It consists of two major modules: 1) an enhanced feature extraction network using successive dilation-deformable feature extraction blocks, which can extract multiscale spatial-spectral features over unfixed sampling locations. It enlarges the field of view of convolutions and takes arbitrary neighborhood into consideration, which helps to increase the discriminativeness of the extracted features; 2) a change sensitive feature augmentation and comparison module integrating feature selection and spatial-channel augmentation strategies, which can exploit spatial context and channel importance. It magnifies difference between changed pixels and unchanged ones and emphasizes contribution of significant channels of the selected change sensitive features. Despite that convolution operations are included in RICD, all the weights are untrained and fixed once they are randomly initialized, indicating that the RICD can work in an unsupervised manner. Its performance is tested over three widely used hyperspectral datasets. Quantitative and qualitative comparisons with several state-of-the-art unsupervised methods reveal the effectiveness of the RICD method." @default.
• W4361801444 created "2023-04-05" @default.
• W4361801444 creator A5022662736 @default.
• W4361801444 creator A5025719998 @default.
• W4361801444 creator A5038354368 @default.
• W4361801444 creator A5040002414 @default.
• W4361801444 creator A5057453193 @default.
• W4361801444 creator A5073447333 @default.
• W4361801444 date "2023-01-01" @default.
• W4361801444 modified "2023-10-05" @default.
• W4361801444 title "From Trained to Untrained: A Novel Change Detection Framework Using Randomly Initialized Models With Spatial–Channel Augmentation for Hyperspectral Images" @default.
• W4361801444 cites W1677182931 @default.
• W4361801444 cites W1964069486 @default.
• W4361801444 cites W2006383776 @default.
• W4361801444 cites W2036798369 @default.
• W4361801444 cites W2104374858 @default.
• W4361801444 cites W2122700624 @default.
• W4361801444 cites W2133059825 @default.
• W4361801444 cites W2134969826 @default.
• W4361801444 cites W2153864221 @default.
• W4361801444 cites W2601564443 @default.
• W4361801444 cites W2804902458 @default.
• W4361801444 cites W2898923688 @default.
• W4361801444 cites W2900587135 @default.
• W4361801444 cites W2902613830 @default.
• W4361801444 cites W2910587630 @default.
• W4361801444 cites W2911648799 @default.
• W4361801444 cites W2953308875 @default.
• W4361801444 cites W2964013315 @default.
• W4361801444 cites W3009942016 @default.
• W4361801444 cites W3015167329 @default.
• W4361801444 cites W3017940382 @default.
• W4361801444 cites W3027201985 @default.
• W4361801444 cites W3044657819 @default.
• W4361801444 cites W3099831940 @default.
• W4361801444 cites W3105553032 @default.
• W4361801444 cites W3129277615 @default.
• W4361801444 cites W3134663792 @default.
• W4361801444 cites W3154634026 @default.
• W4361801444 cites W3164855829 @default.
• W4361801444 cites W3181727745 @default.
• W4361801444 cites W3182928821 @default.
• W4361801444 cites W3183729446 @default.
• W4361801444 cites W3186066750 @default.
• W4361801444 cites W3196356615 @default.
• W4361801444 cites W3197557720 @default.
• W4361801444 cites W3207488093 @default.
• W4361801444 cites W3208491925 @default.
• W4361801444 cites W3215176889 @default.
• W4361801444 cites W3217071007 @default.
• W4361801444 cites W4200206103 @default.
• W4361801444 cites W4205379033 @default.
• W4361801444 cites W4206267281 @default.
• W4361801444 cites W4206550078 @default.
• W4361801444 cites W4211155378 @default.
• W4361801444 cites W4214601364 @default.
• W4361801444 cites W4214836281 @default.
• W4361801444 cites W4221013571 @default.
• W4361801444 cites W4226256612 @default.
• W4361801444 cites W4280554994 @default.
• W4361801444 cites W4281675222 @default.
• W4361801444 cites W4283721824 @default.
• W4361801444 cites W4285111599 @default.
• W4361801444 cites W4285176380 @default.
• W4361801444 cites W4285224173 @default.
• W4361801444 cites W4292825900 @default.
• W4361801444 cites W4293193225 @default.
• W4361801444 cites W4293812296 @default.
• W4361801444 cites W4294691181 @default.
• W4361801444 cites W4294811351 @default.
• W4361801444 cites W4309730479 @default.
• W4361801444 doi "https://doi.org/10.1109/tgrs.2023.3262928" @default.
• W4361801444 hasPublicationYear "2023" @default.
• W4361801444 type Work @default.
• W4361801444 citedByCount "2" @default.
• W4361801444 countsByYear W43618014442023 @default.
• W4361801444 crossrefType "journal-article" @default.
• W4361801444 hasAuthorship W4361801444A5022662736 @default.
• W4361801444 hasAuthorship W4361801444A5025719998 @default.
• W4361801444 hasAuthorship W4361801444A5038354368 @default.
• W4361801444 hasAuthorship W4361801444A5040002414 @default.
• W4361801444 hasAuthorship W4361801444A5057453193 @default.
• W4361801444 hasAuthorship W4361801444A5073447333 @default.
• W4361801444 hasConcept C114614502 @default.
• W4361801444 hasConcept C127162648 @default.
• W4361801444 hasConcept C138885662 @default.
• W4361801444 hasConcept C151730666 @default.
• W4361801444 hasConcept C153180895 @default.
• W4361801444 hasConcept C154945302 @default.
• W4361801444 hasConcept C159078339 @default.
• W4361801444 hasConcept C160633673 @default.
• W4361801444 hasConcept C203595873 @default.
• W4361801444 hasConcept C2776401178 @default.
• W4361801444 hasConcept C2779343474 @default.
• W4361801444 hasConcept C2780757906 @default.
• W4361801444 hasConcept C31258907 @default.
• W4361801444 hasConcept C33923547 @default.
• W4361801444 hasConcept C41008148 @default.

• next