{"622614":{"#nid":"622614","#data":{"type":"news","title":"Georgia Tech Student Leads Creation of the First High-resolution Land Cover Map of the US","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003E\u003Cem\u003EA new deep learning approach enables the first large-scale charting method that can quickly create high-resolution land maps at a fraction of the cost.\u003C\/em\u003E\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EBeing able to predict land cover at a fine resolution from satellite imagery is an extremely important and challenging task in producing key information for many sustainability-related problems.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn an effort to tackle this need on a large scale, a Microsoft research team led by Georgia Tech Ph.D. student\u0026nbsp;\u003Cstrong\u003ECaleb Robinson\u0026nbsp;\u003C\/strong\u003Ecreated the\u0026nbsp;\u003Ca href=\u0022https:\/\/www.youtube.com\/watch?time_continue=8\u0026amp;v=9aFUzUlHQVc\u0022\u003Efirst high-resolution land cover map of the contiguous United States\u003C\/a\u003E.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThis specialized map combines satellite and aerial imagery with deep learning \u0026ndash; a specialized field within artificial intelligence \u0026ndash; to show potentially daily changes in land types such as water, forests, fields, and human developments.\u0026nbsp;\u003Ca href=\u0022https:\/\/www.microsoft.com\/en-us\/ai\/ai-for-earth?activetab=pivot1%3aprimaryr6\u0022\u003EThe Microsoft AI for Earth\u003C\/a\u003E\u0026ndash;funded project will be used in a wide range of applications, particularly by conservation biologists to detect and analyze changes in environments.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;The goal is to identify and summarize changes in land cover on the scale of days,\u0026rdquo; said Robinson. \u0026ldquo;The scale of days is extremely important as this can inform conservation agencies as to how effective their interventions are or this can highlight potential problem areas such as rapid loss of forest cover.\u0026rdquo;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EBistra Dilkina\u003C\/strong\u003E, a\u0026nbsp;\u003Ca href=\u0022https:\/\/www.cse.gatech.edu\/\u0022\u003ESchool of Computational Science and Engineering\u003C\/a\u003E\u0026nbsp;adjunct professor and researcher\u0026nbsp;who specializes in technology for conservation, is a co-investigator on this project and advisor of Robinson\u0026rsquo;s.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026quot;Overall, artificial intelligence has a lot to offer in helping conserve our environment, from land cover mapping and detection of degradation to optimizing conservation planning and predicting risk of illegal wildlife poaching in protected areas,\u0026rdquo; she said.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThese conservation assessments and planning initiatives made possible by high-resolution land maps are important to the health of natural ecosystems and surrounding populations. However, these maps were previously created manually or semi-manually \u0026mdash; an extremely expensive and painstakingly slow method that scales poorly over large areas.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EBy introducing a multi-resolution data fusion method for deep learning\u0026ndash;based high-resolution land cover mapping, the team has successfully reduced the time needed to create this type of map as well as the cost.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;Other deep learning\u0026shy;\u0026ndash;based methods for land cover mapping have shown to be effective but are limited in their scale,\u0026rdquo; said Robinson. \u0026ldquo;By developing methods for generalizing models to new regions, we have achieved high-quality results for the entire US.\u0026rdquo;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EAn example of this cost and time disparity can be seen in the creation of the land cover map of the\u0026nbsp;\u003Ca href=\u0022https:\/\/www.chesapeakebay.net\/what\/maps\/P80?\u0022\u003EChesapeake Bay Watershed in the northeast US\u003C\/a\u003E. The\u0026nbsp;\u003Ca href=\u0022https:\/\/chesapeakeconservancy.org\/\u0022\u003EChesapeake Conservancy\u003C\/a\u003E\u0026nbsp;spent 10 months and $1.3 million to produce the map for an area that measures at approximately 64,299 square miles. Now, with the development of this new method, the team was able to produce a country-wide high-resolution land cover map of 3.20 million square miles within a week for a cost of $5000.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIt also opens the doors to future environmental organizations with limited funding within the US or other areas with commercial satellite imagery providers.\u003C\/p\u003E\r\n\r\n\u003Cp\u003ETo see an example of this map\u0026rsquo;s abilities, Robinson\u0026rsquo;s team partnered with\u0026nbsp;the\u0026nbsp;Chesapeake Conservancy to test this new method by producing a\u0026nbsp;\u003Ca href=\u0022https:\/\/chesapeakeconservancy.org\/conservation-innovation-center\/high-resolution-data\/enhanced-flow-paths\/\u0022\u003Eland cover map for the Middle Cedar Watershed\u003C\/a\u003E\u0026nbsp;in Iowa. This particular map was used to produce high-resolution flow path data sets in order to identify opportunity areas for planting riparian forest buffers \u0026ndash; vegetated spaces that protect streams from collecting pollutants from adjacent areas.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe deep learning method and a detailed account of how the team collaborated with the Middle Cedar Watershed is outlined in a\u0026nbsp;\u003Ca href=\u0022https:\/\/www.cais.usc.edu\/wp-content\/uploads\/2019\/04\/cvpr2019-land-cover-mapping.pdf\u0022\u003Epaper\u003C\/a\u003E, set to be presented at the premier annual computer vision conference,\u0026nbsp;\u003Ca href=\u0022http:\/\/cvpr2019.thecvf.com\/\u0022\u003ECVPR 2019,\u003C\/a\u003E\u0026nbsp;Thursday, June 20.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EAn addition to the map outlined in the paper is a new web interface that allows users to:\u003C\/p\u003E\r\n\r\n\u003Cul\u003E\r\n\t\u003Cli\u003EExplore land cover model predictions\u003C\/li\u003E\r\n\t\u003Cli\u003EFine-tune the model to new geographic areas\u003C\/li\u003E\r\n\t\u003Cli\u003EExtend the model with new, user defined classes\u003C\/li\u003E\r\n\t\u003Cli\u003EDownload resulting land cover data\u003C\/li\u003E\r\n\u003C\/ul\u003E\r\n\r\n\u003Cp\u003EThe code for training and testing these models can be accessed\u0026nbsp;\u003Ca href=\u0022https:\/\/github.com\/calebrob6\/land-cover\u0022\u003Ehere\u003C\/a\u003E.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"A Microsoft team, led by CSE Ph.D. researcher Caleb Robinson, created the first high-resolution land cover map of the contiguous US."}],"uid":"34540","created_gmt":"2019-06-19 18:33:26","changed_gmt":"2019-07-02 16:21:20","author":"Kristen Perez","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2019-06-19T00:00:00-04:00","iso_date":"2019-06-19T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"hg_media":{"622613":{"id":"622613","type":"image","title":"Atlanta Land Cover Map","body":null,"created":"1560968948","gmt_created":"2019-06-19 18:29:08","changed":"1560968948","gmt_changed":"2019-06-19 18:29:08","alt":"Image of a land cover map of Atlanta from a satellite image with labeling.","file":{"fid":"237132","name":"Atlanta Zoom 1.png","image_path":"\/sites\/default\/files\/images\/Atlanta%20Zoom%201.png","image_full_path":"http:\/\/www.tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/Atlanta%20Zoom%201.png","mime":"image\/png","size":1706250,"path_740":"http:\/\/www.tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/Atlanta%20Zoom%201.png?itok=QzTBoCwJ"}}},"media_ids":["622613"],"groups":[{"id":"47223","name":"College of Computing"},{"id":"431631","name":"OMS"},{"id":"50877","name":"School of Computational Science and Engineering"},{"id":"576481","name":"ML@GT"}],"categories":[],"keywords":[],"core_research_areas":[{"id":"39431","name":"Data Engineering and Science"},{"id":"39501","name":"People and Technology"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EKristen Perez\u003C\/p\u003E\r\n\r\n\u003Cp\u003ECommunications Officer\u003C\/p\u003E\r\n","format":"limited_html"}],"email":["kristen.perez@cc.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}