{"componentChunkName":"component---src-templates-blog-post-js","path":"/blog/aerialwaste-dataset-for-landfill-discovery-in-aerial-and-satellite-images/","result":{"data":{"site":{"siteMetadata":{"title":"No Frills News"}},"contentfulNfnPost":{"postTitle":"AerialWaste dataset for landfill discovery in aerial and satellite images","slug":"aerialwaste-dataset-for-landfill-discovery-in-aerial-and-satellite-images","createdLocal":"2023-02-01 14:30:53.570389","publishDate":"None","feedName":"Image Recognition","sourceUrl":{"sourceUrl":"https://www.nature.com/articles/s41597-023-01976-9"},"postSummary":{"childMarkdownRemark":{"html":"<p>The location database is used to extract the positive images from the three data sources.\nA subset of the positive images is enriched with segmentation masks with the help of the ODIN annotator tool.\nWorldView-3 (WV3): high-resolution pan-sharpened RGB images acquired by a commercial satellite sensor (no pan-sharpened near infrared images were used).\nThe evidence attribute is present in most samples, and the severity and area type annotations are specified for ≈72% of the samples.\n5 Distribution of images in the dataset by the (a) evidence, (b) severity and (c) area type properties (n/s = not specified).</p>"}}}},"pageContext":{"slug":"aerialwaste-dataset-for-landfill-discovery-in-aerial-and-satellite-images"}}}