Watch videos with subtitles in your language, upload your videos, create your own subtitles! Click here to learn more on "how to Dotsub"


0 (0 Likes / 0 Dislikes)
The talk is about geographic computation and visualization. Some of us shorten it to geographic computation but I will also talk about visualization. So my name is Bjorn Zimmerman, I am Wolfram Alpha Scientific Content Manager, which means I represent another branch of the Wolfram family. Wolfram Alpha we have made since its launch in 2009 and at that time it makes us actually also available for external users, that’s why we made this initiative with our latest products where we have now abilities to create very nicely and easily maps. We can be found right here—we are right here in the documentation center: geographic data and computation. I will mostly cover, hopefully if time allows us —and if not you can easily read it up later—on maps and cartography, geographic data & entities, and geodetic computations and I probably have no time to talk about this thing here. So essentials—what do we need to do very flexible geographic maps? Well, we have to have something like graphics. Graphics are a very powerful part of the Wolfram Language and can be used in very many ways and one way was we decided to use it now for our geographic visualization. In the Wolfram Language we have also geodetic computation. We have geodetic computation in the Wolfram Language since version 7 so we can already build on that, we don’t have to start from scratch. And with the new version of the Wolfram products the external users also have access to the Wolfram knowledge base. So this means they have access to a lot of entities, which we will use because a lot of entities which we have in Alpha they have positions associated with them, polygons associated with them. We can easily use this now for geographic visualizations. In Wolfram knowledge base, we have streetmap tiles and we have variable elevation data. Geodetic computations. There are many ways to model to Earth depending on what approximation is accessible for you or what you actually want you can choose one of those. But what they all have in common is that they are describing the position of the Earth in latitude and longitude. So that, for marking coordinates in the Wolfram language is GeoPosition. As you might have noticed, we have switched some of the regular paclets, for instance country data, they return now polygons in GeoPosition. This is very nice because it means we now treat things in a uniform way and if the value gives you a polygons GeoPosition then the shipping paclet also gives you things wrapped in GeoPosition. as I said, has latitude and longitude—before it was always the other way around as longitude/latitude because you could easily plot it in graphics. This means graphics also had to learn about GeoPosition, which is just now in the latest version of our products. Since we are now promoting GeoPosition, we have to expand GeoPosition. So now GeoPosition erases arguments; it can autoconvert input; it supports more input forms; it now supports quantities, which we introduced in the last version of Mathematica for external users; using since the launch also Wolfram Alpha; you can also use now DMS strings in GeoPosition or you can geo-tag images. If you use geo-tagged images, you can easily extract information on where the image was taken —that’s information that is in the metadata information of the image. There’s also GeoPositions there some [], which we have some varieties that we also expanded accordingly. GeoProjection data. Again, geoProjection data is for quite some while available for external users in the Wolfram Language. We made various tweaks to tune the projections and many more are coming. GeoDisplacement. GeoDisplacement is very convenient to describe routes or courses. Before you could use, for instance, GeoDistance between position one and position two. Or you can say, “Give me the GeoDistance between a position and the bearing,” which means you go in a certain direction for a certain distance This you can now easily describe in GeoDisplacement. Again, there is a very nice variety. You can easily check—we have very nice documentation, which we also updated quite a bit for our latest release. GeoDistance. GeoDistance was in before; we again, we expanded it and now it supports entities, it supports the distance functions. You can imagine you have two polygons and you want to know the nearest position between these two polygons or the nearest position between the centers of the polygons. All of these things you can easily specify now. You can imagine they often are quite different. And as we do in Wolfram Alpha, many of these are geo-aware functions support now unit system so if you are used to metrics—if you are in a country where the metric system is used— you will probably use meters otherwise you will get imperial unit like miles. We introduce our GeoDistance with extended entities, for instance GeoNearest give me all the countries which are within one hundred kilometers of Germany, you get something like this. GeoIdentify give me all polygons where my current Geoposition, right here, is empty. Where I am part of—we have time zones, all sorts of different time zones. In this case you can see we have polygons for a lot of entities. GeoWithinQ—is New York City within the United States? Yes, it is. Or, for instance, GeoEntities we can say, “Which kinds of buildings do you know about that are in New Orleans?” You can choose whatever you like. Geographic Visualization— one has to be aware that they geographics uses coordinates on two layers, this means before and after the projection. So this means we had to expand the options you are used to, graphics to make a geo alternative for it. Once you can say easily geoposition “Here” and then you get right here in Champaign, so we get some nice area in Champaign, we some notion about where you are. You can click on it, you can click on period, and you can easily see latitude and longitude, where actually are certain positions, you use the map source, you can click on those, you can copy and paste those, and you have already positions you can use for [] different maps. I think this is nicely covered in the documentation. Most of the options are automatic. Once we evaluate it we try to get sensible results. If you want to know this or use this further, you can look at options, you can find a lot of information about which automatic setting was translated to. For instance, we have some meta information, of course we annotate things, or we can say that projection, use Mercator projection, which is very common in these kind of things, in these kind of formats, of small areas. I do not have time to talk about all these options here so if you get the notebook you will find very nice descriptions. Again, we brought very nice documentation— we think it is very nice, hopefully you feel the same. GeoZoomLevel—what kind of data do we actually have now available? As I said, we have our street map tiles, we have from the whole world zoom level 1 and up to the very local location up to zoom level 18. You can see that the system tries to show a sensible GeoRange, details for the range you’ve chosen; it’s kind of sensible. You can do the same now for elevation data, a set we have. With elevation data we have up to 8 at the moment. It is currently in process to add higher resolution elevation data maps. This is coming, just didn’t quite make it into this version. But again, this is background on the Wolfram Alpha site, so in this case we can easily maybe even push these things without having to release a new version of Mathematica. Set GeoProjection—we have a lot of GeoProjections and there are many, many more coming in the next version so we are having some effort doing here. GeoModel—as I said before there are many different models to describe Earth, we have some already available in GeodesyData, for example WGS84, which is used for GPS coordinates. You can use your custom ellipsoid parameters or you can even use entities for which we have good maps available. For instance, for Mars or Moon it can use the very same thing. You can use GeoGridLines, you can plot everything, we did a very nice blog post on that for the Apollo Moon landings so we can see that very nicely. GeoScaleBar is very nice. We say what scale we are talking about, we have some sense of scale for your… I think I’m getting close to the end here. Primitives. We have a bunch of primitives. GeoMarker will be so you can easily annotate maps, we have enough for it again that we a lot of predefined icons available. You can easily mark up your maps. GeoPaths—again, this is paths on the body. So if you use a normal line between two points you’re drilling through the Earth. You don’t want to use this—you want to use GeoPaths to describe now the path on the body. So if you have different versions, like Geodesic or Loxodrome or Rhumb line, these are used in aviation if you want to go in a certain direction for certain distance and you can easily do that. We have named paths like “Meridian”, “Parallel”, “ArticCircle”, “TropicOfCancer'' and again there are some nice examples in here but I do not have time at the moment. GeoGroup—you are able to group primitives as you are able to do with [] , you can do something similar now also with GeoGroup they can do primitives. GeoStyling takes precedence over normal graphics directives, that’s something you may need to get used to but it is very powerful. The non-uniform GeoStyling employs texture —in this case you can really build [] maps. You can use tools tips on it, mouse over every single word nicely and we have predefined some various GeoStyles—outline maps, normal street maps, street map without labels, elevation map, contour map —you can easily do this now. Or you can, as Shadi showed, you can use very nicely also our powerful ImageEffects; there’s the charcoal effect, you can apply this also for your maps. Whatever you can think of. Again, we have very nice documentation on this. We have GeoListPlot and GeoRegionValuePlot which I do not have time, which would be a derivative of that. I think two shorts things, I think I have time for that. For instance, you have a very nice camera which gives off GPS coordinates, taking some pictures and in the end you download your pictures and you think, “Oh where did I take this?” an easy way, say for instance we were walking around here in Champaign or I think it’s Urbana actually, you can see where the pictures were taken. So you can easily do this or you can map your GPS coordinates, I think Chris showed us in the beginning. You can easily gain connection to devices, through connected devices media, and then you can easily do that. We can, again, try to map to Wolfram. Here we are in Champaign—how far are we away? These are the different offices we have for our Wolfram family. As you can see are using Interpreter here, which Jeremy will talk to you about a little bit more later. I want to thank our special teams, so we have Jose, Francisco, who are our team in South America, Eric, which handles a lot of our documentation, and Brett and Michael with a lot of ideas, apps and thank you.

Video Details

Duration: 12 minutes and 37 seconds
Language: English
License: Dotsub - Standard License
Genre: None
Views: 26
Posted by: wolfram on Apr 15, 2015


Caption and Translate

    Sign In/Register for Dotsub to translate this video.