Matches in SemOpenAlex for { <https://semopenalex.org/work/W4280491503> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4280491503 endingPage "312" @default.
- W4280491503 startingPage "292" @default.
- W4280491503 abstract "Convolutional neural networks (CNNs) have achieved extraordinary success on many image classification tasks in recent years. The use of dilated convolution in a CNN can increase the network’s receptive field and improve its performance, and dilated convolution can also be used to compress a CNN to realize a lightweight model. In previous studies, multiscale dilated convolution has been adopted with a focus on improving the internal network structure of a specific CNN model. Because they enable the direct use of pretrained models, transfer learning CNNs (TL-CNNs) have been widely applied for image recognition based on small datasets. This paper proposes a novel multiscale dilated-convolution-based ensemble learning (MDCEL) method for effectively improving the performance of a pretrained CNN model. The authors' primary assumption is that semantic representations of different images can be obtained based on multiscale dilated convolution. Therefore, constructing an ensemble of diverse TL-CNN classifiers makes it possible to achieve higher performance than that offered by the traditional TL-CNN methods. The MDCEL method is highly versatile and can be applied to various conventional pretrained CNN models and lightweight CNN models. Moreover, this method does not require the modification of the internal structures of the pretrained CNN models and has high training efficiency. Experimental results on three public image classification datasets demonstrate that the proposed method outperforms the baseline traditional TL-CNN method. Compared with the baseline approach, the MDCEL approach improves the accuracy and F1 values by nearly 1–4%. In addition, an experiment on a real case dataset obtained from a manufacturing enterprise further proves the practicability of the proposed method." @default.
- W4280491503 created "2022-05-22" @default.
- W4280491503 creator A5017372589 @default.
- W4280491503 creator A5028614906 @default.
- W4280491503 creator A5067246240 @default.
- W4280491503 creator A5089040874 @default.
- W4280491503 date "2022-08-01" @default.
- W4280491503 modified "2023-10-01" @default.
- W4280491503 title "Enhancing ensemble diversity based on multiscale dilated convolution in image classification" @default.
- W4280491503 cites W2031489346 @default.
- W4280491503 cites W2063875787 @default.
- W4280491503 cites W2110764733 @default.
- W4280491503 cites W2611814915 @default.
- W4280491503 cites W2786225488 @default.
- W4280491503 cites W2911033192 @default.
- W4280491503 cites W2915142517 @default.
- W4280491503 cites W2943925420 @default.
- W4280491503 cites W2946875819 @default.
- W4280491503 cites W2960671912 @default.
- W4280491503 cites W2965417721 @default.
- W4280491503 cites W2988698702 @default.
- W4280491503 cites W2997484060 @default.
- W4280491503 cites W3004543888 @default.
- W4280491503 cites W3008497156 @default.
- W4280491503 cites W3012614932 @default.
- W4280491503 cites W3016237374 @default.
- W4280491503 cites W3036760948 @default.
- W4280491503 cites W3089176850 @default.
- W4280491503 cites W3100321043 @default.
- W4280491503 cites W3126968387 @default.
- W4280491503 cites W3131995483 @default.
- W4280491503 doi "https://doi.org/10.1016/j.ins.2022.05.064" @default.
- W4280491503 hasPublicationYear "2022" @default.
- W4280491503 type Work @default.
- W4280491503 citedByCount "8" @default.
- W4280491503 countsByYear W42804915032022 @default.
- W4280491503 countsByYear W42804915032023 @default.
- W4280491503 crossrefType "journal-article" @default.
- W4280491503 hasAuthorship W4280491503A5017372589 @default.
- W4280491503 hasAuthorship W4280491503A5028614906 @default.
- W4280491503 hasAuthorship W4280491503A5067246240 @default.
- W4280491503 hasAuthorship W4280491503A5089040874 @default.
- W4280491503 hasConcept C108583219 @default.
- W4280491503 hasConcept C115961682 @default.
- W4280491503 hasConcept C119857082 @default.
- W4280491503 hasConcept C153180895 @default.
- W4280491503 hasConcept C154945302 @default.
- W4280491503 hasConcept C41008148 @default.
- W4280491503 hasConcept C45347329 @default.
- W4280491503 hasConcept C50644808 @default.
- W4280491503 hasConcept C75294576 @default.
- W4280491503 hasConcept C81363708 @default.
- W4280491503 hasConceptScore W4280491503C108583219 @default.
- W4280491503 hasConceptScore W4280491503C115961682 @default.
- W4280491503 hasConceptScore W4280491503C119857082 @default.
- W4280491503 hasConceptScore W4280491503C153180895 @default.
- W4280491503 hasConceptScore W4280491503C154945302 @default.
- W4280491503 hasConceptScore W4280491503C41008148 @default.
- W4280491503 hasConceptScore W4280491503C45347329 @default.
- W4280491503 hasConceptScore W4280491503C50644808 @default.
- W4280491503 hasConceptScore W4280491503C75294576 @default.
- W4280491503 hasConceptScore W4280491503C81363708 @default.
- W4280491503 hasFunder F4320321878 @default.
- W4280491503 hasFunder F4320322795 @default.
- W4280491503 hasLocation W42804915031 @default.
- W4280491503 hasOpenAccess W4280491503 @default.
- W4280491503 hasPrimaryLocation W42804915031 @default.
- W4280491503 hasRelatedWork W2470368200 @default.
- W4280491503 hasRelatedWork W2766604260 @default.
- W4280491503 hasRelatedWork W2912288872 @default.
- W4280491503 hasRelatedWork W2986507176 @default.
- W4280491503 hasRelatedWork W2996856019 @default.
- W4280491503 hasRelatedWork W3018421652 @default.
- W4280491503 hasRelatedWork W3160224718 @default.
- W4280491503 hasRelatedWork W3160711233 @default.
- W4280491503 hasRelatedWork W4220996320 @default.
- W4280491503 hasRelatedWork W4312417841 @default.
- W4280491503 hasVolume "606" @default.
- W4280491503 isParatext "false" @default.
- W4280491503 isRetracted "false" @default.
- W4280491503 workType "article" @default.