Remote Sensing #18 - ENVI Sentinel 2, download, georeference, layer stack

hey everyone and today I'm gonna quickly show you how to get some settled Landsat images and then you can use them in NV so if you go to the USGS website you can first put in a place for example I can put in something like Sydney and if I hit sure I can find a location and then I want to do is just put in anywhere between two dates or a range of two days and then what I have to do is I have to go to data sets and then I want to bring down Landsat and I can bring down level two which is the better quality and you can also bring down a sentinel as well and then when you have these two things assuming you'll first logged in you can add these few things and then you can go to results and then you can find a bunch of these things and then all you have to do is just go to all the same and then again assuming you're logged in you can order them and then they'll go to your email and when they go to your email you'll get like a table with like a link and assuming it's been approved which usually takes like 10 minutes after you get the initial message you'll have a document coming up something like this and when you go to this you'll be able to scroll all the way down to download and when you download this you'll get a zip file which will give you a bunch of a bunch of jp2 type files so if I just select all these you just do file open to bring them all in and you just open more up and then as soon as you do that you'll realize that every single one of these JP twos is a image of the same area but in different wavelength so you just want to go down to you just want to go down to under radiometric correction and then you want to find often this one off that's not correct underneath radiometric also not under very much kinda roster management you will then find ideal and then layer stacking but just before you use that you have to make sure that all these are corrected so if I tried to do layer stacking to begin with I try to import all these okay you'll say the info file must be G referenced in order to do this you just you just have to right-click on each and every single one of these gray ones each of these is a different color so for example this is B 12 B 11 B 10 these means bands so I've been one I would assume is blue so you have blue green red me and Fred ba-ba-ba-ba-ba all the way up so we'll do Ben one first make sure that you do these in order so you I grew to right-click view metadata and then just edit metadata wait for this to come up and then all you have to do is just hit OK and so what this will do is it'll do reference the image for you so just hit OK and then it'll be moved up to the top so this image is now 0 reference correctly and so you want to do this for each of them and then once you've done it for all of them I'll just do one more even edit metadata okay so assuming that I've done all of them which you have to do you can then go over to IDL and then layer stacking so ever this to load though you're stacking and then you can just import your true that you did so one and two should both be finished yep okay so you'll see that it now works and then you can just output this to some location I'll just do to memory and then you just hit okay and it'll what land stacking is it gets all the different band layers so blue green red blah blah blah and it puts them all into one image so then once you have that you now have a image which you've downloaded you've geo-referenced you put all the colors together and now you can just do regular classification and rear metric Corrections yeah because this itself is essentially geometric Corrections without using crowd control points so yeah that's how you do that and I hope you guys have learned something and I'm curious what these two will come out as okay so it comes out like a grayscale but if you do all them together then you'll get a colored image

Channel: The Rational Planner

Views: 6,048

Keywords: the rational planner, envi, layer stacking, georeferencing

Id: HybyQnFq-sI

Length: 4min 18sec (258 seconds)

Published: Tue Oct 08 2019