Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197956558> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3197956558 endingPage "3562" @default.
- W3197956558 startingPage "3562" @default.
- W3197956558 abstract "Fine-scale sea ice conditions are key to our efforts to understand and model climate change. We propose the first deep learning pipeline to extract fine-scale sea ice layers from high-resolution satellite imagery (Worldview-3). Extracting sea ice from imagery is often challenging due to the potentially complex texture from older ice floes (i.e., floating chunks of sea ice) and surrounding slush ice, making ice floes less distinctive from the surrounding water. We propose a pipeline using a U-Net variant with a Resnet encoder to retrieve ice floe pixel masks from very-high-resolution multispectral satellite imagery. Even with a modest-sized hand-labeled training set and the most basic hyperparameter choices, our CNN-based approach attains an out-of-sample F1 score of 0.698–a nearly 60% improvement when compared to a watershed segmentation baseline. We then supplement our training set with a much larger sample of images weak-labeled by a watershed segmentation algorithm. To ensure watershed derived pack-ice masks were a good representation of the underlying images, we created a synthetic version for each weak-labeled image, where areas outside the mask are replaced by open water scenery. Adding our synthetic image dataset, obtained at minimal effort when compared with hand-labeling, further improves the out-of-sample F1 score to 0.734. Finally, we use an ensemble of four test metrics and evaluated after mosaicing outputs for entire scenes to mimic production setting during model selection, reaching an out-of-sample F1 score of 0.753. Our fully-automated pipeline is capable of detecting, monitoring, and segmenting ice floes at a very fine level of detail, and provides a roadmap for other use-cases where partial results can be obtained with threshold-based methods but a context-robust segmentation pipeline is desired." @default.
- W3197956558 created "2021-09-13" @default.
- W3197956558 creator A5078210453 @default.
- W3197956558 creator A5090461618 @default.
- W3197956558 date "2021-09-08" @default.
- W3197956558 modified "2023-10-16" @default.
- W3197956558 title "Fine-Scale Sea Ice Segmentation for High-Resolution Satellite Imagery with Weakly-Supervised CNNs" @default.
- W3197956558 cites W1976012308 @default.
- W3197956558 cites W1987869189 @default.
- W3197956558 cites W2002101446 @default.
- W3197956558 cites W2002312361 @default.
- W3197956558 cites W2015827190 @default.
- W3197956558 cites W2037497238 @default.
- W3197956558 cites W2054100321 @default.
- W3197956558 cites W2063093906 @default.
- W3197956558 cites W2072189936 @default.
- W3197956558 cites W2117539524 @default.
- W3197956558 cites W2128332833 @default.
- W3197956558 cites W2156739270 @default.
- W3197956558 cites W2754558739 @default.
- W3197956558 cites W2799805314 @default.
- W3197956558 cites W2884166952 @default.
- W3197956558 cites W2900119926 @default.
- W3197956558 cites W2915254566 @default.
- W3197956558 cites W2940726923 @default.
- W3197956558 cites W2974652835 @default.
- W3197956558 cites W3021288641 @default.
- W3197956558 cites W3040593397 @default.
- W3197956558 cites W3099319035 @default.
- W3197956558 cites W3147886766 @default.
- W3197956558 cites W3153376936 @default.
- W3197956558 cites W3157221545 @default.
- W3197956558 cites W3168688383 @default.
- W3197956558 cites W4247505106 @default.
- W3197956558 doi "https://doi.org/10.3390/rs13183562" @default.
- W3197956558 hasPublicationYear "2021" @default.
- W3197956558 type Work @default.
- W3197956558 sameAs 3197956558 @default.
- W3197956558 citedByCount "3" @default.
- W3197956558 countsByYear W31979565582022 @default.
- W3197956558 countsByYear W31979565582023 @default.
- W3197956558 crossrefType "journal-article" @default.
- W3197956558 hasAuthorship W3197956558A5078210453 @default.
- W3197956558 hasAuthorship W3197956558A5090461618 @default.
- W3197956558 hasBestOaLocation W31979565581 @default.
- W3197956558 hasConcept C108583219 @default.
- W3197956558 hasConcept C127313418 @default.
- W3197956558 hasConcept C136894858 @default.
- W3197956558 hasConcept C150547873 @default.
- W3197956558 hasConcept C153180895 @default.
- W3197956558 hasConcept C154945302 @default.
- W3197956558 hasConcept C205649164 @default.
- W3197956558 hasConcept C2778102629 @default.
- W3197956558 hasConcept C2778755073 @default.
- W3197956558 hasConcept C31972630 @default.
- W3197956558 hasConcept C41008148 @default.
- W3197956558 hasConcept C49204034 @default.
- W3197956558 hasConcept C58640448 @default.
- W3197956558 hasConcept C62649853 @default.
- W3197956558 hasConcept C89600930 @default.
- W3197956558 hasConceptScore W3197956558C108583219 @default.
- W3197956558 hasConceptScore W3197956558C127313418 @default.
- W3197956558 hasConceptScore W3197956558C136894858 @default.
- W3197956558 hasConceptScore W3197956558C150547873 @default.
- W3197956558 hasConceptScore W3197956558C153180895 @default.
- W3197956558 hasConceptScore W3197956558C154945302 @default.
- W3197956558 hasConceptScore W3197956558C205649164 @default.
- W3197956558 hasConceptScore W3197956558C2778102629 @default.
- W3197956558 hasConceptScore W3197956558C2778755073 @default.
- W3197956558 hasConceptScore W3197956558C31972630 @default.
- W3197956558 hasConceptScore W3197956558C41008148 @default.
- W3197956558 hasConceptScore W3197956558C49204034 @default.
- W3197956558 hasConceptScore W3197956558C58640448 @default.
- W3197956558 hasConceptScore W3197956558C62649853 @default.
- W3197956558 hasConceptScore W3197956558C89600930 @default.
- W3197956558 hasIssue "18" @default.
- W3197956558 hasLocation W31979565581 @default.
- W3197956558 hasLocation W31979565582 @default.
- W3197956558 hasOpenAccess W3197956558 @default.
- W3197956558 hasPrimaryLocation W31979565581 @default.
- W3197956558 hasRelatedWork W1669643531 @default.
- W3197956558 hasRelatedWork W2005437358 @default.
- W3197956558 hasRelatedWork W2008656436 @default.
- W3197956558 hasRelatedWork W2039154422 @default.
- W3197956558 hasRelatedWork W2122581818 @default.
- W3197956558 hasRelatedWork W2517104666 @default.
- W3197956558 hasRelatedWork W2790662084 @default.
- W3197956558 hasRelatedWork W2948658236 @default.
- W3197956558 hasRelatedWork W4293211451 @default.
- W3197956558 hasRelatedWork W2182382398 @default.
- W3197956558 hasVolume "13" @default.
- W3197956558 isParatext "false" @default.
- W3197956558 isRetracted "false" @default.
- W3197956558 magId "3197956558" @default.
- W3197956558 workType "article" @default.