I have classified (supervised classification) sentinel-2 optical imagery using google earth engine java-script api. In one project i have also used python api of Google earth engine for calculation indices NDVI, EVI etc. Now i want to use the same(google earth engine python api) for image classification. But i am unable to figure out how i can provide training sets for classification from earth engine python api. Is it possible to save the training points in fusion table and use that for training in python api for GEE? or is there any other way to do the same.
Advance thanks for your assistence.
You can export to KML (Export.table.toDrive()), then upload to Fusion Table or export to Earth Engine asset (Export.table.toAsset()) and read it directly from there.
Related
I'm trying to visualize GPS points on PowerBI maps offline and wonder what would be the best way to achieve that.
Ideally, I'd like to see the offline mapping to have the same effect as directly using the "Map" visualization - GPS points (in data as longitude and latitude columns) plotted on the Bing map. However, our team will need this function offline, because our users are usually under bad internet connections. Also we are wary about Bing's data sharing policy as our GPS information is highly confidential.
So far I've tried PowerBI's shape map function, which allows me to upload a TopoJSON file and use it offline. However, it looks like shape map is better used for highlighting a region on the map, instead of plotting individual GPS points - shape map doesn't seem to allow longitude/latitude. Besides that, shape map does not have as much information like Bing map, such as showing city/road names and etc.
Is there a way to visualize GPS points offline on PowerBI, but still get enough information about what are around the GPS points? I'm referring to PowerBI here but all things I've been doing are on PowerBI Desktop. I don't plan to publish on the web as this will be my company's internal tool. Thanks in advance!
The built-in mapping functions such as the Map visual powered by the Bing maps API and the Eris mapping do not have an offline mode. Both require an online connection. Depending on the level of detail that you require, I would suggest looking into an R or Python visual with one of the mapping libraries. Here are some examples of the R mapping libraries. But getting the map json files into Power BI will also be an issue and may not meet your requirements on the level of detail that you need.
Also we are wary about Bing's data sharing policy as our GPS information is highly confidential.
I wouldn't worry about it, Bing & Power BI mapping is used by a wide number of organisation including heath and military customers. Your confidential data is more at risk with some one copying your Power BI report and sticking it on a USB drive.
However, our team will need this function offline, because our users are usually under bad internet connections
I would try some other options for offline mapping, for example Google Maps has the ability to down load a section of the map for offline/bad connections. I've downloaded the whole UK and its about 100meg.
Hope that helps
I am trying to use HERE Maps to implement the eco-routing algorithm which we developed in our lab that requires real time traffic data as an input to the routing algorithm. So far, we were using INRIX dataset to verify our algorithms, but we decided to move to HERE Maps, and use the Traffic API to harness the power of real time traffic data.
Traffic API was very simple to use, and I managed to get the traffic data for a bounding box. It seems like that the traffic flow is passed for each TMC. My question is how to assign the traffic flow data to the links of the map using TMC. More specifically, I am wondering if there is a way to transform the TMC values to an adjacency matrix so that I can assign the traffics to each link and run my optimization algorithm over that graph? Also is it possible to overlay Open Street Maps (OSM) with the HERE Maps traffic data?
Thanks
You can associate the TMC data you get from the traffic data response to a particular link by using PDE layers LINK_TMC_FC1..5. You have look into our PDE documentation in developer site and tcs.ext.here.com/pde/maps?url_root=pde.api.here.com to see the list of layers provided.
Below is a sample PDE request for the Link TMC. Replace app id and code placeholders with actuals:
https://pde.api.here.com/1/tile.json?&layer=LINK_TMC_FC1&level=9&tilex=496&tiley=358®ion=WEU&release=18131&url_root=pde.api.here.com&app_id={YOUR-APP-ID}&app_code={YOUR-APP-CODE}
For all your other questions on creating a matrix and OSM overlay, you have to write your own algorithm. There is no straight forward function provided by Here.
I want download multiple Google Earth images randomly at chosen locations within the study area. Each image had a spatial extent of 250m×250m which corresponds, or exceeds the size of a pixel from medium spatial resolution satellite sensors (e.g. MODIS or Landsat),how to do this in R, and how to save these images?
I would suggest reading the Terms of Service for Maps and Earth API.
Paragraph 10.1 (a) says
No access to APIs or Content except through the Service. You will not access the Maps API(s) or the Content except through the Service. For example, you must not access map tiles or imagery through interfaces or channels (including undocumented Google interfaces) other than the Maps API(s).
That means you cannot download and save the imagery on your side, it will be considered as violation of ToS.
I would like to visualize into maps results of MDX queries on spatial dimensions (representing provinces,clusters,neighborhoods...) stored in Postgis.
I found about the GeoMap component, but I don't know if it's the right tool for this purpose.
How do I plot latitude and logitude using Google Geocoder webservice in asp.net?
There is a full tutorial on how to use Google maps in java script, so if you are using asp.net to display a web page, it isn't too hard:
https://developers.google.com/maps/documentation/javascript/tutorial
If you are trying to plot lat and long on to a custom made maps, you would have to deal with the fact that the earth is basically a sphere, and do 3D modeling. Its quite math intensive, and this probably isn't the correct place for this question.