Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386880059> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W4386880059 abstract "Abstract Industrial materials images are an important application domain for content-based image retrieval. Users need to quickly search databases for images that exhibit similar appearance, properties, and/or features to reduce analysis turnaround time and cost. The images in this study are 2D images of millimeter-scale rock samples acquired at micrometer resolution with light microscopy or extracted from 3D micro-CT scans. Labeled rock images are expensive and time-consuming to acquire and thus are typically only available in the tens of thousands. Training a high-capacity deep learning (DL) model from scratch is therefore not practicable due to data paucity. To overcome this “few-shot learning” challenge, we propose leveraging pretrained common DL models in conjunction with transfer learning. The “similarity” of industrial materials images is subjective and assessed by human experts based on both visual appearance and physical qualities. We have emulated this human-driven assessment process via a physics-informed neural network including metadata and physical measurements in the loss function. We present a novel DL architecture that combines Siamese neural networks with a loss function that integrates classification and regression terms. The networks are trained with both image and metadata similarity (classification), and with metadata prediction (regression). For efficient inference, we use a highly compressed image feature representation, computed offline once, to search the database for images similar to a query image. Numerical experiments demonstrate superior retrieval performance of our new architecture compared with other DL and custom-feature-based approaches." @default.
- W4386880059 created "2023-09-21" @default.
- W4386880059 creator A5018028862 @default.
- W4386880059 creator A5035621021 @default.
- W4386880059 creator A5051892293 @default.
- W4386880059 creator A5055768448 @default.
- W4386880059 creator A5074462430 @default.
- W4386880059 date "2023-01-01" @default.
- W4386880059 modified "2023-09-27" @default.
- W4386880059 title "Content-based image retrieval for industrial material images with deep learning and encoded physical properties" @default.
- W4386880059 cites W1953465585 @default.
- W4386880059 cites W2019702606 @default.
- W4386880059 cites W2042650924 @default.
- W4386880059 cites W2117539524 @default.
- W4386880059 cites W2132201434 @default.
- W4386880059 cites W2194775991 @default.
- W4386880059 cites W2476063773 @default.
- W4386880059 cites W2499468060 @default.
- W4386880059 cites W2755957465 @default.
- W4386880059 cites W2774740309 @default.
- W4386880059 cites W2791358116 @default.
- W4386880059 cites W2794284562 @default.
- W4386880059 cites W2888436193 @default.
- W4386880059 cites W2944083325 @default.
- W4386880059 cites W2971102415 @default.
- W4386880059 cites W2972825732 @default.
- W4386880059 cites W2978481357 @default.
- W4386880059 cites W3080427797 @default.
- W4386880059 cites W3108274592 @default.
- W4386880059 cites W3133795864 @default.
- W4386880059 cites W3136841107 @default.
- W4386880059 cites W3156930162 @default.
- W4386880059 cites W3181472034 @default.
- W4386880059 cites W4223601593 @default.
- W4386880059 cites W4229450040 @default.
- W4386880059 cites W4282557376 @default.
- W4386880059 doi "https://doi.org/10.1017/dce.2023.16" @default.
- W4386880059 hasPublicationYear "2023" @default.
- W4386880059 type Work @default.
- W4386880059 citedByCount "0" @default.
- W4386880059 crossrefType "journal-article" @default.
- W4386880059 hasAuthorship W4386880059A5018028862 @default.
- W4386880059 hasAuthorship W4386880059A5035621021 @default.
- W4386880059 hasAuthorship W4386880059A5051892293 @default.
- W4386880059 hasAuthorship W4386880059A5055768448 @default.
- W4386880059 hasAuthorship W4386880059A5074462430 @default.
- W4386880059 hasBestOaLocation W43868800591 @default.
- W4386880059 hasConcept C103278499 @default.
- W4386880059 hasConcept C108583219 @default.
- W4386880059 hasConcept C111919701 @default.
- W4386880059 hasConcept C115961682 @default.
- W4386880059 hasConcept C119857082 @default.
- W4386880059 hasConcept C138885662 @default.
- W4386880059 hasConcept C150899416 @default.
- W4386880059 hasConcept C153180895 @default.
- W4386880059 hasConcept C154945302 @default.
- W4386880059 hasConcept C1667742 @default.
- W4386880059 hasConcept C2776214188 @default.
- W4386880059 hasConcept C2776401178 @default.
- W4386880059 hasConcept C41008148 @default.
- W4386880059 hasConcept C41895202 @default.
- W4386880059 hasConcept C50644808 @default.
- W4386880059 hasConcept C59404180 @default.
- W4386880059 hasConcept C93518851 @default.
- W4386880059 hasConceptScore W4386880059C103278499 @default.
- W4386880059 hasConceptScore W4386880059C108583219 @default.
- W4386880059 hasConceptScore W4386880059C111919701 @default.
- W4386880059 hasConceptScore W4386880059C115961682 @default.
- W4386880059 hasConceptScore W4386880059C119857082 @default.
- W4386880059 hasConceptScore W4386880059C138885662 @default.
- W4386880059 hasConceptScore W4386880059C150899416 @default.
- W4386880059 hasConceptScore W4386880059C153180895 @default.
- W4386880059 hasConceptScore W4386880059C154945302 @default.
- W4386880059 hasConceptScore W4386880059C1667742 @default.
- W4386880059 hasConceptScore W4386880059C2776214188 @default.
- W4386880059 hasConceptScore W4386880059C2776401178 @default.
- W4386880059 hasConceptScore W4386880059C41008148 @default.
- W4386880059 hasConceptScore W4386880059C41895202 @default.
- W4386880059 hasConceptScore W4386880059C50644808 @default.
- W4386880059 hasConceptScore W4386880059C59404180 @default.
- W4386880059 hasConceptScore W4386880059C93518851 @default.
- W4386880059 hasFunder F4320333062 @default.
- W4386880059 hasLocation W43868800591 @default.
- W4386880059 hasOpenAccess W4386880059 @default.
- W4386880059 hasPrimaryLocation W43868800591 @default.
- W4386880059 hasRelatedWork W2546942002 @default.
- W4386880059 hasRelatedWork W2945839551 @default.
- W4386880059 hasRelatedWork W2960456850 @default.
- W4386880059 hasRelatedWork W2970216048 @default.
- W4386880059 hasRelatedWork W3192840557 @default.
- W4386880059 hasRelatedWork W3195938642 @default.
- W4386880059 hasRelatedWork W4312200629 @default.
- W4386880059 hasRelatedWork W4317565044 @default.
- W4386880059 hasRelatedWork W4382286161 @default.
- W4386880059 hasRelatedWork W4386213806 @default.
- W4386880059 hasVolume "4" @default.
- W4386880059 isParatext "false" @default.
- W4386880059 isRetracted "false" @default.
- W4386880059 workType "article" @default.