{"id":115,"date":"2024-07-10T20:45:01","date_gmt":"2024-07-10T20:45:01","guid":{"rendered":"https:\/\/www.ssec.wisc.edu\/leads\/?page_id=115"},"modified":"2026-04-20T18:44:31","modified_gmt":"2026-04-20T18:44:31","slug":"leads-detection-using-ai","status":"publish","type":"page","link":"https:\/\/www.ssec.wisc.edu\/leads\/leads-detection-using-ai\/","title":{"rendered":"Leads Detection using AI"},"content":{"rendered":"\n

The full description of the AI based sea ice leads detection algorithm can be found in Application of a Convolutional Neural Network for the Detection of Sea Ice Leads<\/a> and A 20-Year Climatology of Sea Ice Leads Detected in Infrared Satellite Imagery Using a Convolutional Neural Network<\/a>.<\/p>\n\n\n\n

Below, a small 512 km \u00d7 512 km sample of the VIIRS 11 \u03bcm brightness temperature image from 0024 UTC on 1 January 2020 (warm features are bright) in panel (A<\/strong>). The corresponding U-Net greyscale image of segmentation results in panel (B<\/strong>). Where the segmentation results are above the detection threshold, the binary lead labels are shown in red in panel (C<\/strong>).<\/p>\n\n\n\n

\"Example<\/figure>\n\n\n\n

Below, the relative frequency of detection score for all points (A<\/strong>), for leads in the validation mask (B<\/strong>), and for non-lead locations (C<\/strong>) in the validation masks. This is for the aggregate of 1 January 2020, 1 February 2020, 1 March 2020, and 1 April 2020. The detection thresholds are shown as black lines at 45 for VIIRS and 32 for MODIS.<\/p>\n\n\n\n

\"Example<\/figure>\n\n\n\n

Below n the left, the receiver operating characteristic (ROC) curve for MODIS (red) and VIIRS (blue) shows the detection threshold at the intersection of the 10% false positive rate (with a true positive rate of 97% for VIIRS and MODIS) is selected as the detection threshold. On the right, the same true positive and false positives curves are shown, but this time as a function of detection score.<\/p>\n\n\n\n

\"ROC<\/figure>\n\n\n\n

Below, the VIIRS (top) and MODIS (bottom) 11 \u03bcm composite brightness temperature imagery and leads product results from (left to right) 1 January 2020, 1 February 2020, 1 March 2020, and 1 April 2020. Blue is true negative, green is true positive, red is false positive, and black is false negative in comparison with the validation mask.<\/p>\n\n\n\n

\"4<\/figure>\n","protected":false},"excerpt":{"rendered":"

The full description of the AI based sea ice leads detection algorithm can be found in Application of a Convolutional Neural Network for the Detection of Sea Ice Leads and A 20-Year Climatology of Sea Ice Leads Detected in Infrared Satellite Imagery Using a Convolutional Neural Network. Below, a small 512 km \u00d7 512 km […]<\/p>\n","protected":false},"author":42,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"footnotes":""},"class_list":["post-115","page","type-page","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/pages\/115","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/users\/42"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/comments?post=115"}],"version-history":[{"count":2,"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/pages\/115\/revisions"}],"predecessor-version":[{"id":131,"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/pages\/115\/revisions\/131"}],"wp:attachment":[{"href":"https:\/\/www.ssec.wisc.edu\/leads\/wp-json\/wp\/v2\/media?parent=115"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}