From DEM to point cloud

if you add received a raster a digital elevation model you may extract from it a point cloud so you can go from a two-dimensional raster information each pixel has got information about the coordinates the x-coordinate and y-coordinate of its center and information about the elevation of X each of these cells and you can transform that in a point cloud so you can transform in a data which is made by points each point is positioned in the center of the cells of the raster and it's got information about the coordinates and it's got information about elevation everything is taken from the roster I'm gonna use ku GIS software which is super open soft open source software that allows you to do a lot of things and I'll go through some of these features in some next videos and then I will import the results this point cloud inside cut compare before starting I'd like just to point out a couple of things so the density of this point cloud that we will have in output it's extremely related and linked to the resolution of the raster file so if you have a raster file that is that is got revolution a resolution of 1 meters which is the case that I'm gonna use in this video you will expect to have an output a dense point cloud that is made by points that are equally distributed in the two-dimensional space and they the distance between these points is more or less 1 meters it's far it's extremely it's so different from the data that you may have from a structure from motion processing or a laser scanning acquisition that really that that really represent the 3d 3 the 3d mode that the the 3d aspect of what you have surveyed you are linked to the raster if you get a raster with a resolution of 1 centimeters 4 pixel you have a more dense cloud than the raster with a resolution of 10 meters 4 pixel that's it and this point will be equally distributed in the two-dimensional space so let's jump inside QGIS let's start this processing there will be extremely fast this is the raster digital elevation model well this is a digital terrain model DTM but it's not that's not the point you can do this processing just with the digital elevation model digital terrain model digital surface model whatever model you have that works of course you you need to know what you are extracting from this model if there are points of the ground or if they are a point of the ground plus points of some trees or buildings if i zoom in I can see the roster I'm sorry I think that I should have colored it in a different way but I think maybe here ok okay here's the raster the resolution is 1 meter if I measure this length it will be 1 meter if you'd like to go from the raster to the point cloud you need to open up the processing tools so processing toolbox and I'm going to delete this one thing that I strongly advise you to do and this applies to everything you need to do inside QGIS is just typing in in this search box what you are trying to do what you'd like to do with the plug-in of QGIS and probably there will be something that fits that suit for your purposes so now I'd like to go from a raster file to a point file some got just type inning raster to point and the software filtered me out some results and here we go I've got some options here there are two plugins that basically do the same thing at raster values to point which is a plug-in inside the saga plug-in to box saga G's plug-in to box orders this plug-in whereas there's something also in in the grass toolbox but there's something here in the gujiyas plug-in which is inside a vector creation box and this raster pixel two points is the plug-in that I'm gonna use for this purpose so double click it's extremely easy the software is asking me for is asking you for the digital elevation model the raster that is going to use for this processing there's just one raster in this project if you add more than one you add the options for choosing them choosing one of them just one so click here band number gray field name okay let's step back and choose a different name there's no there's no point in doing that but just in order to let you see what will happen when the point cloud is being generated let's let me close this one I've got this tool here active which is the identify features I'm gonna select the DTM here and I'm just clicking here in the one of this roster on the one of this cell here I got information about the coordinate x coordinate y coordinate these are projected coordinate it's so important having projecting coordinate because when you will go when you will upload this new point cloud inside software that AE deep point cloud such as Cal compare I will do that in cloud compare you will need to have projected coordinates you will not have you don't want to have angles latitude and longitude mix up with meters or length which is the elevation so this is projected coordinate system these are the coordinates and this is the band number one this is the elevation information you can have more than one band one information here it's up to the database file connected to this information in this case we just have this information the of the elevation and I'm gonna change this number this field from value which is the name that is associated by default in this plugin to the name that I'd like to have raster pixels to point maybe I would like to change this one in elevation which suits better field name elevations so this plug-in will create a field name for each point that has got the elevation associated to the elevation field name and this will help us in cloud compare when we will upload this data inside it vector points you need to choose if you'd like to create temporary file save to your package or save to file you need to choose the folder and you need to choose the file name so I'm gonna close I'm gonna call it points file type shapefile but you got also a lot of options here and it's up to you I'm gonna use a shapefile because I know that it will work in most of the software that I'm using for point cloud editing safe once you've done it click run and the software will start the processing I hope that everything will go out fine without any problems in my recording and that's it once done just complete closed and the vector file has been added to the project as you can see the software created one point for each center point of each cells if I select it and I'm just click on on it here we go there's the derived information which is coordinate X Y and then this field that the field that has been created the elevation field which is the elevation information that the software the cloud compare will take in account and it will upload this data inside its its brain ensign its workspace here we go as you can see these points are evenly distributed over the surface cover by this digital elevation model so now let's go inside cloud compare and let's upload this data to see what happens here we go the software QGIS created three new files which basically refers to the same thing but a shapefile is always linked to some other files but I just need to keep take this SHP file and bring inside cloud compare that's a good thing because when I am uploading it inside called compare the software is asking me for global shifting that means that the software realizes that this data is made up by coordinates that are big numbers so is asking me for shifting in order to better deal with this data and that's that's a good thing because the geo referencing information as kept as being kept inside this this file so yes to all the siRNA numbers that have been important here we go now the software cloud compare is asking me which is the altitude field so he he was able to recognize an information about the candle field which is the elevation field that we just typed in a couple of minutes ago and it is asking me what is the information about the elevation if I ignore this every point will be flattened out so there will not will be no information about the elevation if I click here and just click OK the software will give to each of this point the information about the vision so this is the point cloud uploaded in cloud compare of course in the top view it sounds a little bit weird but if I boom remove it and it's upside down alright if I move it there's the elevation that's the point cloud of course this is a digital terrain model so you can see the differences of elevation but it's not so clear and they're not so that's the same thing as if I had some buildings here because I removed this building and of course here there are some weird points just because the software that I use for the building the creation of the digital elevation model was was forced to work in a rectangular surface so he modeled me out this point properly and he modeled me out these borderline points assigning them an elevation of minus 30 meters so these points of course have no points these are points that aren't there in the real world so I can get rid of them inside Cal compare extremely fast I'm gonna move in the side view select them select the point cloud and I'm gonna segment out okay and that's much better this is the these are the points that I was able to to create from the digital elevation model so that's increase the defaults default point size as you can see they are evenly distributed on the surface covered by the digital elevation model let me click on the point picking tool and let's just ask for the coordinates of this point here we go these are the coordinates of this point local coordinate x coordinate y coordinate Z coordinate locals these are the coordinate that has been recalculated the reprocess by the software once I said yes to the global shifting at the very beginning of the uploading processing these are the global coordinates so the coordinates the real the real coordinates of these points will the elevation that has been kept inside the software this the point ups sorry and of course there is no color information there is no color information I do not have the chance for forgiving this now at this point some information about RGB because nothing is related in terms of RGB information cameras or something like that so there's no color information but anyway this is a nice data this is a nice weight for bringing a 2d file a raster file that has information about your referencing and elevation inside and transform it not bring in transforming it in a point cloud which can be used for many purposes it can be used as a kind of a basis for other processing volume computations things like that and that's that that's it I think that's a nice tool and it's extremely fast of course if you have a huge roster it will take a longer time if you have a roster with high resolution it will take a longer time but it works and I think that it's really nice tool to know and to to use it for your works so I hope you enjoyed this video that's it for today feel free to share it feel free to connect with me at www.mathantics.com my name is Paulo Ragini thank you for watching see you in the next one ciao

Channel: Metrical Talks

Views: 396

Rating: 4.6923075 out of 5

Keywords: qgis, 2d, 3d, from 3d to 2d, from 2d to 2d, open source, open source software, dem, dtm, dsm, point cloud, raster

Id: w2n2efNeTpE

Length: 14min 34sec (874 seconds)

Published: Thu Jun 04 2020