{"id":332593,"date":"2017-02-07T03:27:59","date_gmt":"2017-02-06T19:27:59","guid":{"rendered":"http:\/\/techcrunch.com\/?p=1448588"},"modified":"2017-02-07T03:27:59","modified_gmt":"2017-02-06T19:27:59","slug":"descartes-labs-opens-its-geospatial-analysis-engine-to-a-handful-of-lucky-developers","status":"publish","type":"post","link":"https:\/\/people.utm.my\/asmawisham\/descartes-labs-opens-its-geospatial-analysis-engine-to-a-handful-of-lucky-developers\/","title":{"rendered":"Descartes Labs opens its geospatial analysis engine to a handful of lucky developers"},"content":{"rendered":"<p><img decoding=\"async\" src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=738\" class=\"\"\/><\/p>\n<p>It\u2019s easy to forget that even with the fanciest of\u00a0machine learning models, we still need humans in the trenches cleaning input data. <a target=\"_blank\" href=\"http:\/\/www.descarteslabs.com\">Descartes Labs<\/a>,\u00a0a startup that combines satellite imagery with data about our planet to produce insights and forecasts, knows this all too well. The company ended up building\u00a0its own\u00a0cloud-based parallel computing infrastructure to clean and process its massive\u00a0corpus of satellite\u00a0imagery.\u00a0Today it\u2019s\u00a0giving\u00a0a\u00a0handful of developers and early customers access to <a target=\"_blank\" href=\"http:\/\/www.descarteslabs.com\/maps.html\">this\u00a0system.<\/a><\/p>\n<p>Companies like\u00a0Descartes Labs cannot just throw\u00a0raw satellite imagery into machine learning models to extract insights. Images captured contain clouds, cloud shadows and other atmospheric aberrations that make it impossible to compare images taken at different times. A small cloud over a field, for example, that wasn\u2019t present\u00a0in previous images,\u00a0could completely throw off a model attempting to predict crop yields.<\/p>\n<p>To overcome this challenge, engineers can use composite images to optimize for the best pixels across a collection of images. Google Maps\u00a0employs composite imagery to remove clouds and create representations of the globe that are evenly lit by the sun.<\/p>\n<p>The problem with combining dozens of satellite captures of the entire earth is that it\u2019s incredibly computationally intensive. This is where Descartes Labs\u2019s processing engine comes into play to convert\u00a0 into composites the petabytes of\u00a0geospatial data it has.<\/p>\n<div class=\"slideshow slideshowify\">\n<ol readability=\"1\">\n<li readability=\"1\">\n<div class=\"image\"><img loading=\"lazy\" decoding=\"async\" data-thumb=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-11.jpg?w=80&#038;h=60&#038;crop=1\" data-full-size-image=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-11.jpg\" data-large-size-image=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-11.jpg?w=1&#038;h=1\" data-trigger-notification=\"1\" data-scalable=\"false\" alt=\"\" data-src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-11.jpg?w=1&#038;h=1\" data-tc-lazyload=\"deferred\" src=\"https:\/\/s0.wp.com\/wp-content\/themes\/vip\/techcrunch-2013\/assets\/images\/1x1.png\" width=\"1\" height=\"1\"\/><\/div>\n<h3 class=\"title\">unnamed-1<\/h3>\n<p>Raw pixels collected, which include clouds, shadows and haze<\/p>\n<\/li>\n<li readability=\"-1\">\n<div class=\"image\"><img loading=\"lazy\" decoding=\"async\" data-thumb=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-2.jpg?w=80&#038;h=60&#038;crop=1\" data-full-size-image=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-2.jpg\" data-large-size-image=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-2.jpg?w=1&#038;h=1\" data-trigger-notification=\"1\" data-scalable=\"false\" alt=\"\" data-src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-2.jpg?w=1&#038;h=1\" data-tc-lazyload=\"deferred\" src=\"https:\/\/s0.wp.com\/wp-content\/themes\/vip\/techcrunch-2013\/assets\/images\/1x1.png\" width=\"1\" height=\"1\"\/><\/div>\n<h3 class=\"title\">unnamed-2<\/h3>\n<p>Same image atmospherically corrected with clouds and shadows removed<\/p>\n<\/li>\n<li readability=\"-1\">\n<div class=\"image\"><img loading=\"lazy\" decoding=\"async\" data-thumb=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=80&#038;h=60&#038;crop=1\" data-full-size-image=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg\" data-large-size-image=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=1&#038;h=1\" data-trigger-notification=\"1\" data-scalable=\"false\" alt=\"\" data-src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=1&#038;h=1\" data-tc-lazyload=\"deferred\" src=\"https:\/\/s0.wp.com\/wp-content\/themes\/vip\/techcrunch-2013\/assets\/images\/1x1.png\" width=\"1\" height=\"1\"\/><\/div>\n<h3 class=\"title\">unnamed-3<\/h3>\n<p>Composite of best pixels from 24 processed images<\/p>\n<\/li>\n<\/ol><\/div>\n<p>In the images above, you can contrast the low quality of an initial capture with the high quality of a composite \u2014 \u00a0the latter being far better for feeding into deep learning models. To get from start to finish,\u00a0Descartes used\u00a0mathematical techniques to convert the light in the first image to a certain base level before\u00a0masking out clouds.\u00a0After finishing that basic cleaning, the team then had to formally create the\u00a0composite.\u00a0Descartes turned the problem into a\u00a0massive parallel compute job.<\/p>\n<p>The result? The\u00a0first-ever composites created using images captured by the\u00a0Sentinel-1 and 2A satellites. Descartes grouped\u00a0these composites with a third made from images captured by the LandSat 8 to create an online tool for exploring traditional imagery alongside unique\u00a0Synthetic Aperture Radar (SAR) and red-edge imagery. Descartes uses a variety of satellite bands to monitor vegetation\u00a0and other changes happening on the earth\u2019s surface.<\/p>\n<p>One major shortcoming of composite images is that they\u2019re hard to contrast against each other over short time intervals. Because multiple captures have to be combined, they can really only be used for\u00a0comparison across months or quarters. This still has value, particularly when thinking about the financial world\u2019s reliance on the quarter system, but it makes it hard to use for week-to-week agricultural predictions.<\/p>\n<p>Descartes is limiting the group with access to the underlying infrastructure for the time being. But\u00a0anyone can spend time this afternoon examining <a target=\"_blank\" href=\"https:\/\/obscure-maps.descarteslabs.com\/?layer=sentinel-1_v2_vv_2014-2017#lat=37.7884704&#038;lng=237.5949412&#038;zoom=13.9999\">the\u00a0Sentinel and LandSat composites<\/a> the team created.<\/p>\n","protected":false},"excerpt":{"rendered":"<p><img loading=\"lazy\" decoding=\"async\" width=\"680\" height=\"680\" src=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=680\" class=\"attachment-large size-large wp-post-image\" alt=\"Composite of best pixels from 24 processed images\" srcset=\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=680 680w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=1360 1360w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=150 150w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=300 300w, https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/02\/unnamed-3.jpg?w=768 768w\" sizes=\"(max-width: 680px) 100vw, 680px\" \/>&nbsp;It&#8217;s easy to forget that even with the fanciest of\u00a0machine learning models, we still need humans in the trenches cleaning input data. Descartes Labs,\u00a0a startup that combines satellite imagery with data about our planet to produce insights and forecasts, knows this all too well. The company ended up building\u00a0its own\u00a0cloud-based parallel computing infrastructure to clean and process&hellip; <a href=\"https:\/\/techcrunch.com\/2017\/02\/06\/descartes-labs-opens-its-geospatial-analysis-engine-to-a-handful-of-lucky-developers\/?ncid=rss\">Read More<\/a><\/p>\n<div class=\"feedflare\">\n<a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=QHEX_YGyfnA:5QJhMvrkctI:2mJPEYqXBVI\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=2mJPEYqXBVI\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=QHEX_YGyfnA:5QJhMvrkctI:7Q72WNTAKBA\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=7Q72WNTAKBA\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=QHEX_YGyfnA:5QJhMvrkctI:yIl2AUoC8zA\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=yIl2AUoC8zA\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=QHEX_YGyfnA:5QJhMvrkctI:-BTjWOF_DHI\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?i=QHEX_YGyfnA:5QJhMvrkctI:-BTjWOF_DHI\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=QHEX_YGyfnA:5QJhMvrkctI:D7DqB2pKExk\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?i=QHEX_YGyfnA:5QJhMvrkctI:D7DqB2pKExk\" border=\"0\"><\/img><\/a> <a href=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?a=QHEX_YGyfnA:5QJhMvrkctI:qj6IDK7rITs\"><img decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~ff\/Techcrunch?d=qj6IDK7rITs\" border=\"0\"><\/img><\/a>\n<\/div>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"http:\/\/feeds.feedburner.com\/~r\/Techcrunch\/~4\/QHEX_YGyfnA\" height=\"1\" width=\"1\" alt=\"\"\/> <\/p>\n","protected":false},"author":5817,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"categories":[25],"tags":[59,66,26],"class_list":["post-332593","post","type-post","status-publish","format-standard","hentry","category-technology","tag-media","tag-techcrunch","tag-technology"],"_links":{"self":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/332593","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/users\/5817"}],"replies":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/comments?post=332593"}],"version-history":[{"count":0,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/332593\/revisions"}],"wp:attachment":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/media?parent=332593"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/categories?post=332593"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/tags?post=332593"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}