{"id":97967,"date":"2019-08-06T16:30:06","date_gmt":"2019-08-06T16:30:06","guid":{"rendered":"https:\/\/news.microsoft.com\/?p=433938"},"modified":"2019-08-06T16:30:06","modified_gmt":"2019-08-06T16:30:06","slug":"how-ai-is-helping-track-endangered-species","status":"publish","type":"post","link":"https:\/\/sickgaming.net\/blog\/2019\/08\/06\/how-ai-is-helping-track-endangered-species\/","title":{"rendered":"How AI is helping track endangered species"},"content":{"rendered":"

The Hawaiian<\/span> <\/span><\/b>po\u02bbo-uli<\/span>,<\/span> <\/span>a small<\/span> bird<\/span> from the honeycreeper family<\/span>,<\/span> <\/span>was <\/span>first discovered in 1973<\/span><\/a>. Less than half a century later, it disappeared <\/span>from<\/span> the planet.<\/span> <\/span><\/p>\n

Declared extinct in 2018, it is one of almost 700 vertebrate species that h<\/span>a<\/span>ve<\/span> been<\/span> <\/span>driven to extinction<\/span><\/a> in the last 500 years. According to a <\/span>U<\/span>nited Nations<\/span> report<\/span><\/a> issued earlier this year<\/span> to policymakers, <\/span>one<\/span> million species are at<\/span> risk of extinction:<\/span> <\/span>H<\/span>uman actions threaten more plants and animals than ever before.<\/span> Although the precise number of species on the planet<\/span> is difficult to calculate<\/span>, recent estimates put it <\/span>at <\/span>around 8.7 million<\/span><\/a>.<\/span> <\/span><\/p>\n

The plight of the <\/span>po\u02bbo-uli<\/span> illustrates a key problem for conservationists: <\/span>It is difficult<\/span> to track and monitor populations<\/span> of endangered species<\/span>. Despite efforts to tag and locate the dwindling <\/span>po\u02bbo-uli<\/span> <\/span>population, scientists were unable to unite breeding pairs.<\/span> <\/span><\/p>\n

Now, <\/span>technology is offering hope to scientists. Collecting better data and <\/span>analyzing<\/span> it more effectively with <\/span>machine learning and AI<\/span><\/a> allows conservationists to <\/span>make <\/span>more targeted and timely interventions.<\/span> <\/span> <\/span><\/p>\n

Here are five ways Microsoft and conservation efforts are coming together <\/span>to <\/span>help endangered species.<\/span> <\/span><\/p>\n

[Subscribe to <\/span><\/i>Microsoft on the Issues<\/span><\/i><\/a> for more on the topics that matter most.]<\/span><\/i><\/p>\n

Identifying lions<\/span><\/b> and giraffes<\/span><\/b> <\/span><\/p>\n

\"Lion<\/p>\n

Facial<\/span>\u2013<\/span> and pattern<\/span>\u2013<\/span>recognition technology allows researchers to pinpoint and track individual animals, helping <\/span>scientists<\/span> monitor populations and <\/span>their <\/span>migrations<\/span>.<\/span> <\/span><\/p>\n

However, <\/span>w<\/span>ithout <\/span>the <\/span>distinctive markings <\/span>of <\/span>cheetahs and leopards, it is <\/span>harder <\/span>for researchers to identify individual <\/span>lions<\/span>.<\/span> The Lion Identification Network of Collaborators uses facial recognition techniques to <\/span>monitor population levels<\/span><\/a>.<\/span> Identification involves taking a high-quality photo and zooming in on areas such as whisker spots \u2013 something a <\/span>computer is more capable<\/span><\/a> of than the human eye. Th<\/span>is<\/span> tech<\/span>nology<\/span> helps reduce reliance on expensive and difficult<\/span>\u2013<\/span>to<\/span>\u2013<\/span>fit GPS monitors,<\/span> which can have <\/span>limited <\/span>battery <\/span>life.<\/span> <\/span><\/p>\n

Similar technology is being used by r<\/span>esearchers from Penn State<\/span> University<\/span> and <\/span>the <\/span>Wild Nature Institute<\/span> <\/span>to <\/span>study<\/span> the<\/span> births, deaths<\/span> <\/span>and movements of more than 3,000 giraffes in northern Tanzania<\/span>. <\/span> <\/span><\/p>\n

Habitat loss and <\/span>i<\/span>llegal <\/span>hunting<\/span> for meat ha<\/span>ve<\/span> meant giraffe populations have experienced a dramatic decline<\/span> in recent years<\/span>. Pattern<\/span>\u2013<\/span>recognition technology trained <\/span>to identify the unique markings<\/span><\/a> on a giraffe\u2019s torso can now <\/span>process images in minutes<\/span>, a task<\/span> that previously <\/span><\/a>took <\/span>conservationists weeks.<\/span> <\/span><\/p>\n

Monitoring the impact of humans on grizzly bear populations <\/strong><\/p>\n

Clayton Lamb, a researcher at the University of Alberta<\/span>,<\/span> Canada, is using Microsoft Azure tools along with AI to<\/span> create a <\/span>comprehensive analysis<\/span><\/a> of the human and environmental factors limiting grizzly bear density in British Columbia<\/span>. <\/span> <\/span><\/p>\n

With human populations increasingly expanding into areas of wilderness, the grizzly is coming under threat as its habitat is <\/span>shrinking<\/span>. The<\/span>se<\/span> roaming omnivores are particularly sensitive to human disturbance<\/span>, and globally now <\/span>live<\/span> <\/span>in<\/span> a<\/span>n area <\/span>only <\/span>a<\/span>bout<\/span> <\/span>half <\/span>their<\/span> original range<\/span><\/a>. <\/span> <\/span><\/p>\n

Lamb<\/span> uses DNA sampling and GPS collars to track individual bears <\/span>in order to <\/span>better understand the connectivity between populations and their habitats at a granular level. <\/span>By generating such specific data, he is able to move his work beyond theoretical population ecology and suggest steps to reduce the impact of humans on wildlife populations.<\/span><\/p>\n

Assessing whale health and feeding patterns<\/strong> <\/span><\/p>\n

At Duke University<\/span> in North Carolina<\/span>,<\/span> scientists are using<\/span> <\/span>marine robotics and remote<\/span>\u2013<\/span>sensing<\/span>,<\/span><\/a> machine<\/span>\u2013<\/span>learning models on Microsoft Azur<\/span>e<\/span> to assess data relating to the <\/span>size and health of whales<\/span><\/a>.<\/span> <\/span><\/p>\n

As oceans warm, krill stocks \u2013 a critical whale food \u2013 are falling, while fishing and commercial exploitation of habitats <\/span>are<\/span> also placing a strain on whale populations.<\/span> <\/span><\/p>\n

Given the <\/span>distance<\/span>s<\/span> whales travel, monitoring their movements using traditional methods is both time-consuming and expensive. Neural networks have allowed scientists to link multiple systems<\/span>,<\/span> including satellites, drones and autonomous underwater vehicles<\/span>,<\/span> to paint a more comprehensive picture of the migration and health of these great animals.<\/span><\/p>\n

Tracking penguin populations<\/span><\/b> <\/span><\/p>\n

\"Penguins<\/p>\n

M<\/span>onitoring penguin populations<\/span> in remote habitats<\/span> is <\/span>also <\/span>difficult<\/span> <\/span>work<\/span>. Identifying colonies often involves looking for hard-to-find markers <\/span>such as <\/span>guano stains on rock<\/span>s of a very similar color.<\/span> <\/span> <\/span><\/p>\n

E<\/span>cologist Heather J<\/span>.<\/span> Lynch<\/span>, from<\/span> the Department of Ecology and Evolution<\/span>,<\/span> Stony Brook University<\/span>,<\/span> <\/span>is coupling AI capabilities with predictive population modelling to allow <\/span>real-time tracking of Antarctic penguins<\/span><\/a>.<\/span> <\/span><\/p>\n

She\u2019s using <\/span>c<\/span>omputer vision <\/span>to <\/span>search for guano stains in satellite imagery <\/span>in order <\/span>to <\/span>inform and develop <\/span>the <\/span>classification algorithms <\/span>that<\/span> generate population estimates for colonies. <\/span> <\/span><\/p>\n

Protecting elephants from poachers<\/span><\/b> <\/span><\/p>\n

It is estimated that an <\/span>elephant was killed <\/span>every 15 minutes<\/span><\/a> between 2007 and 2014<\/span> \u2013<\/span> <\/span>a<\/span>nd <\/span>that <\/span>25,000 to 35,000 wer<\/span>e<\/span> killed each year<\/span> by poachers<\/span> <\/span>for their ivory<\/span>.<\/span> <\/span> <\/span> <\/span> <\/span><\/p>\n

With huge areas to monitor, defending <\/span>these animals<\/span> is a challenge<\/span>.<\/span> <\/span>R<\/span>esources <\/span>can<\/span> be thinly spread, communication <\/span>can be<\/span> patchy and gathering usable data tricky.<\/span> <\/span><\/p>\n

The <\/span>Elephant Listening Project<\/span><\/a> is helping protect the threatened forest elephant.<\/span> Teams at Cornell University<\/span> in Ithaca, N<\/span>ew York<\/span>, with <\/span>assistance <\/span>from Conservation Metrics<\/span> <\/span>based in Santa Cruz, C<\/span>alifornia<\/span>, <\/span>are <\/span>using AI to identify and analyze<\/span><\/a> recordings of elephant calls, helping to create an <\/span>\u201c<\/span>elephant dictionary<\/span>\u201d<\/span> by <\/span>matching noises to behavior<\/span><\/a>. <\/span> <\/span><\/p>\n

A<\/span> <\/span>growing<\/span> understanding of elephant communication<\/span> is allowing researchers to better appreciate the impact of factors <\/span>such as <\/span>oil exploration, logging<\/span> and<\/span> poaching<\/span>, as well as <\/span>enabling them<\/span> to<\/span> <\/span>coordinate conservation efforts<\/span> more efficiently<\/span><\/a>.<\/span> <\/span><\/p>\n

For more on how AI can be used to solve global environmental challenges, visit <\/span><\/i>AI for Earth<\/span><\/i><\/a>. And follow @MSFTIssues<\/a> on Twitter.\u202f\u202f<\/span><\/i> <\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

The Hawaiian po\u02bbo-uli, a small bird from the honeycreeper family, was first discovered in 1973. Less than half a century later, it disappeared from the planet.  Declared extinct in 2018, it is one of almost 700 vertebrate species that have been driven to extinction in the last 500 years. According to a United Nations report issued earlier this year to policymakers, one million species are at risk of extinction: Human actions threaten more plants […]<\/p>\n","protected":false},"author":2,"featured_media":97968,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[49],"tags":[135,50],"class_list":["post-97967","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-microsoft-news","tag-artificial-intelligence","tag-recent-news"],"_links":{"self":[{"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/posts\/97967","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/comments?post=97967"}],"version-history":[{"count":0,"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/posts\/97967\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/media\/97968"}],"wp:attachment":[{"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/media?parent=97967"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/categories?post=97967"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sickgaming.net\/blog\/wp-json\/wp\/v2\/tags?post=97967"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}