The mission to create a searchable database of Earth's surface | Will Marshall

Prevodilac: Ivana Korom Lektor: Aleksandar Korom

Four years ago, here at TED,

Pre četiri godine, ovde na TED-u,

I announced Planet's Mission 1:

najavio sam Planetinu Misiju 1:

to launch a fleet of satellites

da lansiramo flotu satelita

that would image the entire Earth, every day,

koja će snimati celu Zemlju, svakog dana,

and to democratize access to it.

i da pristup tome demokratizujemo.

The problem we were trying to solve was simple.

Problem koji smo pokušavali da rešimo bio je jednostavan.

Satellite imagery you find online is old, typically years old,

Satelitski snimci koje nalazite na netu su stari, godinama stari,

yet human activity was happening on days and weeks and months,

a ljudska aktivnost se dešava danima, nedeljama, mesecima,

and you can't fix what you can't see.

a ne možete popraviti ono što ne vidite.

We wanted to give people the tools to see that change and take action.

Želeli smo da ljudima damo alate da vide tu promenu i preduzmu nešto.

The beautiful Blue Marble image, taken by the Apollo 17 astronauts in 1972

Predivna slika Plave planete koju su uslikali astronauti Apola 17, 1972.

had helped humanity become aware that we're on a fragile planet.

pomogla je čovečanstvu da postane svesno da se nalazimo na krhkoj planeti.

And we wanted to take it to the next level,

A mi smo želeli to da prenesemo na sledeći nivo,

to give people the tools to take action, to take care of it.

da damo ljudima alate da preduzmu nešto, da vode računa o njoj.

Well, after many Apollo projects of our own,

Pa, posle mnogih sopstvenih Apolo projekata,

launching the largest fleet of satellites in human history,

lansiranja nаjveće flote satelita u ljudskoj istoriji,

we have reached our target.

dostigli smo svoj cilj.

Today, Planet images the entire Earth, every single day.

Danas, Planeta snima celu Zemlju svakog dana.

Mission accomplished.

Misija ostvarena.

(Applause)

(Aplauz)

Thank you.

Hvala.

It's taken 21 rocket launches --

Bilo je potrebno 21 lansiranje raketa -

this animation makes it look really simple -- it was not.

na ovoj animaciji deluje jednostavno, ali nije bilo.

And we now have over 200 satellites in orbit,

Sada u orbiti imamo preko 200 satelita,

downlinking their data to 31 ground stations we built around the planet.

koji šalju svoje podatke na 31 stanicu koju smo sagradili na zemlji.

In total, we get 1.5 million 29-megapixel images of the Earth down each day.

Ukupno, dobijamo 1,5 milion slika Zemlje, od 29 megapiksela, svaki dan.

And on any one location of the Earth's surface,

I sada imamo više od 500 slika

we now have on average more than 500 images.

bilo koje tačke Zemljine površine.

A deep stack of data, documenting immense change.

Ogromna gomila podataka, koja beleži velike promene.

And lots of people are using this imagery.

Mnogo ljudi koristi ove slike.

Agricultural companies are using it to improve farmers' crop yields.

Poljoprivredne kompanije ih koriste da poboljšaju prinose njiva.

Consumer-mapping companies are using it to improve the maps you find online.

Kompanije koje prave mape koriste ih da poboljšaju mape koje nalazite na netu.

Governments are using it for border security

Vlade ih koriste za pogranično obezbeđenje

or helping with disaster response after floods and fires and earthquakes.

ili da pomažu u odgovorima na nepogode posle poplava, požara ili zemljotresa.

And lots of NGOs are using it.

I mnoge nevladine organizacije ih koriste.

So, for tracking and stopping deforestation.

Za praćenje i zaustavljanje seče šuma.

Or helping to find the refugees fleeing Myanmar.

Ili za pomoć pri pronalaženju izbeglica koje beže iz Mjanmara.

Or to track all the activities in the ongoing crisis in Syria,

Ili da prate sve aktivnosti u tekućoj krizi u Siriji,

holding all sides accountable.

i da sve strane drže odgovornim.

And today, I'm pleased to announce Planet stories.

A danas, imam zadovoljstvo da najavim Planetine priče.

Anyone can go online to planet.com

Svako može da ode na planet.com,

open an account and see all of our imagery online.

napravi nalog i vidi sve naše slike onlajn.

It's a bit like Google Earth, except it's up-to-date imagery,

Liči na Google Earth, osim što su slike ažurirane

and you can see back through time.

i možete da se vraćate kroz vreme.

You can compare any two days

Možete da uporedite bilo koja dva dana

and see the dramatic changes that happen around our planet.

i vidite dramatične promene koje se dešavaju širom planete.

Or you can create a time lapse through the 500 images that we have

Ili možete da napravite video sa 500 slika koje imamo

and see that change dramatically over time.

i vidite dramatičnu promenu kroz vreme.

And you can share these over social media.

A možete i da ih podelite putem društvenih medija.

It's pretty cool.

Prilično je kul.

(Applause)

(Aplauz)

Thank you.

Hvala.

We initially created this tool for news journalists,

Prvobitno smo ovo napravili za novinare,

who wanted to get unbiased information about world events.

koji su želeli da imaju nepristrasne informacije o svetskim događajima.

But now we've opened it up for anyone to use,

Ali sad smo ih otvorili za sve,

for nonprofit or personal uses.

za neprofitno i za lično korišćenje.

And we hope it will give people the tools to find and see the changes on the planet

I nadamo se da će dati ljudima alate da pronađu i vide promene na zemlji

and take action.

i preduzmu nešto.

OK, so humanity now has this database of information about the planet,

OK, sada čovečanstvo ima ovu bazu informacija o planeti

changing over time.

koja se menja kroz vreme.

What's our next mission, what's Mission 2?

Šta je naša sledeća misija, Misija 2?

In short, it's space plus AI.

Ukratko, to je svemir + veštačka inteligencija.

What we're doing with artificial intelligence

Sa veštačkom inteligencijom

is finding the objects in all the satellite images.

pronalazimo predmete u svim satelitskim slikama.

The same AI tools that are used to find cats in videos online

Isti alati veštačke inteligencije koji se koriste da nađu mačke u snimcima,

can also be used to find information on our pictures.

mogu da se koriste da pronađu informacije na našim slikama.

So, imagine if you can say, this is a ship, this is a tree,

Dakle, zamislite da možete da kažete ovo je brod, ovo je drvo,

this is a car, this is a road, this is a building, this is a truck.

ovo je auto, ovo je put, ovo je zgrada, ovo je kamion.

And if you could do that for all of the millions of images

I da možete to da uradite za sve milione slika

coming down per day,

koje se učitavaju dnevno,

then you basically create a database

onda u stvari stvarate bazu podataka

of all the sizable objects on the planet, every day.

svih predmeta na zemlji, svakog dana.

And that database is searchable.

I ta baza je pretraživa.

So that's exactly what we're doing.

To je upravo ono što radimo.

Here's a prototype, working on our API.

Evo prototipa, koji radi s našim API-em.

This is Beijing.

Ovo je Peking.

So, imagine if we wanted to count the planes in the airport.

Dakle, zamislite da želimo da prebrojimo avione na aerodromu.

We select the airport,

Odaberemo aerodrom,

and it finds the planes in today's image,

i pronalazimo avione na današnjoj slici,

and finds the planes in the whole stack of images before it,

i pronalazimo avione na čitavoj gomili slika pre toga,

and then outputs this graph of all the planes in Beijing airport over time.

a onda dobijamo ovaj grafikon svih aviona na pekinškom aerodromu kroz vreme.

Of course, you could do this for all the airports around the world.

Naravno, ovo možete da uradite za sve aerodrome na svetu.

And let's look here in the port of Vancouver.

Hajde da pogledamo ovde, u luci Vankuvera.

So, we would do the same, but this time we would look for vessels.

Možemo da uradimo isto, ali ovog puta tražimo plovila.

So, we zoom in on Vancouver, we select the area,

Dakle, zumiramo na Vankuver, odaberemo oblast,

and we search for ships.

i tražimo brodove.

And it outputs where all the ships are.

I dobijemo prikaz gde su svi brodovi.

Now, imagine how useful this would be to people in coast guards

Sad, zamislite kako bi ovo bilo korisno ljudima u obalskoj straži

who are trying to track and stop illegal fishing.

koji pokušavaju da pronađu i zaustave ilegalno ribarenje.

See, legal fishing vessels

Vidite, brodovi legalnih ribara

transmit their locations using AIS beacons.

odašilju svoju lokaciju preko AIS odašiljača.

But we frequently find ships that are not doing that.

Ali često nalazimo brodove koji to ne rade.

The pictures don't lie.

Slike ne lažu.

And so, coast guards could use that

Tako bi obalske straže mogle da koriste to

and go and find those illegal fishing vessels.

i pronađu te ilegalne ribarske brodove.

And soon we'll add not just ships and planes

Uskoro ćemo dodati ne samo brodove i avione,

but all the other objects,

nego i druge predmete,

and we can output data feeds

i možemo izbacivati hrpu podataka

of those locations of all these objects over time

o tim lokacijama svih tih predmeta tokom vremena,

that can be integrated digitally from people's work flows.

koji mogu da se digitalno integrišu iz podataka ljudi.

In time, we could get more sophisticated browsers

Vremenom, mogli bismo da dobijemo sofisticiranije pretraživače

that people pull in from different sources.

koje ljudi povlače iz raznih izvora.

But ultimately, I can imagine us abstracting out the imagery entirely

Ali na kraju, mogu da zamislim da potpuno apstrahujemo slike

and just having a queryable interface to the Earth.

i samo imamo pretraživ interfejs Zemlje.

Imagine if we could just ask,

Zamislite da možete da pitate:

"Hey, how many houses are there in Pakistan?

"Koliko kuća ima u Pakistanu?

Give me a plot of that versus time."

Izbaci mi nacrt kroz vreme."

"How many trees are there in the Amazon

"Koliko drveća ima u Amazonu

and can you tell me the locations of the trees that have been felled

i daj mi lokacije drveća koje je oboreno

between this week and last week?"

u poslednjih nedelju dana."

Wouldn't that be great?

Zar to ne bi bilo sjajno?

Well, that's what we're trying to go towards,

Pa, ka tome pokušavamo da idemo,

and we call it "Queryable Earth."

i nazivamo ga "Pretraživa Zemlja".

So, Planet's Mission 1 was to image the whole planet every day

Dakle, Planetina Misija 1 je bila da snimimo celu planetu svaki dan

and make it accessible.

i da slike budu dostupne.

Planet's Mission 2 is to index all the objects on the planet over time

Planetina Misija 2 je da popišemo sve objekte na planeti tokom vremena

and make it queryable.

i da budu pretraživi.

Let me leave you with an analogy.

Ostaviću vas sa jednom analogijom.

Google indexed what's on the internet and made it searchable.

Gugl je indeksirao sve na internetu i omogućio da se pretražuje.

Well, we're indexing what's on the Earth and making it searchable.

Mi popisujemo sve što je na Zemlji i omogućavamo da se pretražuje.

Thank you very much.

Mnogo vam hvala.

(Applause)

(Aplauz)