 course on remote sensing and GIS for rural development. This is week 6, lecture 3. In this week, we have been looking at different coordinate system projections and looking at how to collect data from other sources. The projections and coordinate system is also important while downloading vector and raster data, especially from drones and satellites. However, there are already many maps that can be scanned and used for GIS purpose. The focus is also more on the rural development aspect. In the last class, we looked at this database for buying or accessing data. So let us once again revisit this website and download and show an optimum data set that we can use. So I'm opening it in a particular screen. So you could see that when you open the link that I gave, you will come to the online portals directly to the products. And you can see all products visualized here. We looked at what each product is, digital vector database, administrative boundaries, digital integration model, or DEM. Here it's called digital terrain model. And then you have geo reference color raster. So yeah, then you also have digital geographical maps. And each one is a particular attribute. We can go ahead and look at it. So one is the roadmap, India and adjacent countries, world map of India, railway roads of India, et cetera. But you have to buy 5,500. And there's all the rules and regulations on how you can share or use it. Then you go to Village Boundary Database, which is free of cost. You have geo database and shape file. Database full of lot of data or just a shape file you can buy. And then we come to Open Series Free PDF Map, which we are going to use for this. Some maps online, it will say 77 rupees, 100 rupees. But when you go to the buy map section, it will say that there's no cost. So when I click on the, click to buy, you get this image or this page where all of India is divided into grids. There is one sheet that is available for a particular grid. And then you have to zoom in for the number of the type. So there's a sheet and a type. So I'll just ask you what is the number? Enter the sheet number, et cetera. So an example is that man L12, let us zoom in and see if we could look at some of them. So now you could see J43, J44 is there. And then there is ABCD. And within the ABCD there's four small, smaller grids. So each location is divided into multiple grids and there is a naming scheme. So there is J and then on the top, there is some new numbers to show which numbers you want to go. It is kind of difficult to just looking at this map and then collect that data. So for that they have given you a help window for taking out the data. Let us go and visit it. So here go to search sheet number. If you click it, this page will come where you can select a state. And we are going to do Tamil Nadu, for example. And then just wait for it. And then you can go to selects district. Let us say Chennai. And then you have all the different numbers. So you can see here, open street map is D4404. So there is a location and within that there's 1234 divided into multiple OSN numbers and 11 sheet number is there for it. Again, you don't have to worry about how these numberings have been done, but you can zoom in and collect the data. So once you have the OSN number or sheet number, you can press select to place an order. And then there it is, it is selected. Just give it a second so that this gets populated here the sheet number and then you can download. So when you click download, you will have to log in and then mobile number, password, et cetera, et cetera. I would prefer not to put it on my numbers here. So, but still just straight forward, you sign in, you register and then, or if you don't have an account, please register and download the image, okay? So you're going back to the main page where all the data is there and then you can buy. Now, if you click on this since it's already been there, it'll ask you the same questions of which tile number, et cetera, et cetera, you can have. Okay, the boundaries are pretty accurate from here, so please only use these boundaries. You can sign in here or create a login here based on your need. Let's go back to the slide that we were presenting. Yes, so moving on, the next part is extracting geo-referencing points, okay? So for extracting the data from a paper map or a toposheet that we just saw, we need to put points on the map with locations, okay? So these locations should be known by the user so that the paper map can be anchored. Without this, the paper map can float any location. So to make it anchored, you need to have at least four to six points depending on the method that you're using to anchor the map. So we'll see what does that mean, what does that mean, okay? So without a proper location, the map can go anywhere, okay? So that is where you need to anchor, like ship can go anywhere, but it stands because it's anchored. So that is the same word we're using here. And then since we're using maps, maps already have latitude and longitude. So it is optimum for us to use lat-longs, okay? So we could use the data that is already in the map, populate it and then use it for the lat-longs. Zoom in as much as possible to find the data. We will work on this part when we do a hands-on zoom. Examine clearly for using the data for attributes. So first step is to collect the points on the map that is going to anchor your map and then put locations for those points and then make sure that you zoom in as much as possible to get the accurate location and then examine clearly for using the data for attributes. So then the last point is once the map is projected, you can go zoom in and collect the information that is needed. So the georeferencer tool manual is given in this link, which is on the QGIS web page. I'll open it now for the just going through on what it is. So what we will see here is we will go to the QGIS referencing tool and then look at the manual that has QGIS tutorials. We have already used it in the previous versions, right? In the documentation, we went and looked at the modules. And then, yeah. So let me open this link. I've opened it now. I'm going to share the screen. Yes, yeah, now it's visible. So what we will see is that we will see under the manual at the desktop user guide manual 3.222. You have 16.3 is georeferencer. The usual procedure is entering ground control points, GCPs. As I said, you find locations and those are the ground control. The anchor points are called GCPs and then you can see here some other tools that are in the toolbox. And then you have defining the transformation system. What are the different transformation systems that can apply the lat-longs based on the six points? So you have a image frame. So in the image frame, I'll just show like a paper example. In the image, let's say the whole image is in the center, but you have a full A4 size map. So what you'll do is you will just anchor four or five points along the paper and throughout the paper, it will interpolate. So that is what throughout the paper map, it will interpolate and then it becomes digitalized in your model. So that part is known as transformation and then define the resampling method if you need and then show and adapt properties, configure georeferencer, running the transformation. So please go through this manual. I will definitely do hands-on soon on this so that you will know what is the difference in these properties and methods. Here's where you would enter the data and you could enter the data as per the different formats, DMS, DD, MMSS, which is degrees, minutes and seconds, or DD as DD, DD, DD, just degrees, degrees and the symbols or project coordinates, MMM. So mostly the maps have DMS version which is degree, minutes and seconds. So you look at the transforming part. The multiple algorithms to transform is just interpolation methods, but then once it does, it will give you the running after you run the transformation. It will give you the errors based on the model and how the model has performed. Let us go back to the slide. So we have looked into this georeferencer manual and now we will look into the toolbox just to look at what are the different tools that are available. So I will show my QGIS part. I'm opening my QGIS now. Yes, now it's open. So where would you find it? You'll find it in the raster toolbox georeferencer tool. If you click georeferencer, sometimes it'll open and it'll be at the bottom. So just click open, it'll open. You can do it in two steps, which is basically you can have the map that you want inside in the layers and then bring it here as an image or you could do it without the georeferencing. So before that, we want to give a normal indicator of what this data frame is going to be. So I'm going to click open and then I'm going to add full states, my database. And the EPSG was as per the previous WGS84. That looks good. So I'll just keep it. So the datum and the coordinate system is WGS84, EPSG4326. There are multiple coordinate systems you can see here. You click, you can just filter based on India. So there's multiple zones, right? State, Indian. So just look at how many you have, but mostly people use 4326 EPSG or WGS84 and that is what we will use. So there's multiple WGS and then in that you'll have different EPSGs. So we have used one based on the data set that we already have. So we have the India full states. So let it be here for some time. Now what we will do is we're going to input the data that we downloaded from a survey of India. I'm going to open the raster database. This is the tile that I've done. So it says open raster data space and then add. So once we add it, we don't know, it doesn't link on here. So what have I downloaded? I have downloaded a map of Karnataka, Bangalore, very specific. You can see here, we discussed this map in the previous lecture. What are the objects, et cetera, et cetera. I've downloaded this part, which is D43R1257G-12 and that is Bangalore urban, but here we have Bangalore rural. So we want to cover the rural part on the north. So the north part is where we will focus more. However, we will look into the entire tile. So we have the entire tile. And as I said, it is not sitting. It is not anchoring with the location. It's just floating somewhere. You could see here, if I do this, India is this and then this is a tile. You know for sure that India is much, much bigger than the Bangalore map here. So it is technically wrong, right? So for example, I'll show you if I zoom in, this is India, okay? So why does it do it? Because it doesn't have a location. So I'm going to remove this layer and then zoom in again. Now you have only this layer. Let me add the raster again. So it's the rural dot there and then I'm going to close and then I'm going to zoom back to the layers and you can see almost the same effect. India is small and the layer is big. So I will remove this layer for now and then zoom to the full layer of India and let's keep it there for now. The georeferencing tool is up. As I said, here you can add the raster. This is the open raster button. This is the running the georeferencing tool. So right now we will not run it until and otherwise we have put the points and then here are the points, GCP points or anchor points that we need to put it, okay? So first step is to add the image. So I'm going to go to my database, add the image. The image has come, okay? So as I said, there are some features in the map that will give you the lat-long location. So let us look at that. So if you zoom in the top, you can see here it is 77 degrees is this line and then 13 degrees is this line. What is it? 13 degrees, 15 minutes. Whereas this is 77 degrees, 30 minutes. So every grid is spaced at 2 degree, 30 minutes space. So 77, 30 degrees, then this is 32, 30 degrees. If you can see 32, 30 degrees and then we have 35. So 230, 230, 230. Again, 30 plus 30 is 60, 60 is equal to one. It is like same like your clock thing. Minutes, seconds convert to minutes. Minutes convert back to your degrees. So here what you could see is you have 77 degrees, 30 minutes and then 32 minutes, 30 seconds. So in between you have two minute, 30 seconds, okay? And then a two minute, 30 seconds plus it becomes 35 and then two minutes, 30, 37 degrees, 37 minutes, 30 cents. 40, 42, 30 and 45. Same way down, we have 13, 50 and then you can use a hand tool to pull down. You can see 13, 15 down, 12, 30 and another minus 230 you become 10, 7, 30 and then five, 230 and then 13, zero. So this line is 13, zero and this line is 77, 30 because it's the same line, it didn't change but then this line has decreased from the top. So you have 13, 230, now 13, zero. So now let's select some points and I'll show you how to select. So these are the tools that are there. You can add a point or you can import from other sources. Since you're learning how to do this, I'm going to just add a GCP point and you can delete it or move it based on if the point you put is error. The best I would say is just delete it and put it again, moving is not correct. So first step is to identify a point where you want to location where you put a point and then zoom in to put the point. So let's do the easiest one is on the top. So let's take six points, two on the top, two on the bottom and then two in between. So let's go at this point. So I'm going to take 77, 30 line. So this is the 77, 39 and then I'm going to use the 13, 15. So this line is going to be 13, 15 throughout but I'm going to shift here and take a point. The first point, let's take 77, 35. So 77, 35 is this 77 degrees, 35 minutes. As I said, you have to zoom in. You can zoom in using your mouse. You can or this point. Let's do the mouse first. If you zoom in, you see that the black line is kind of smudged because the resolution is getting bad when you zoom in. The black becomes gray, gray becomes white, et cetera. But as you know, two lines when it cross, it is darker there. So you see this black pixel is dark. So you have to put a point there because this represents your 35 line and then here 77, 35, line 77 degrees, 35 minutes. And on this side, it represents 13 degrees, 15 minutes. So I'm going to zoom in to that pixel. You can see the pixel and put the point in the center. This will add more accuracy to your model. So you see how this, you can eyeball or if you want, you can take your tape and measure and put but the model will adjust for these kinds of small errors. Here, the type of the coordinate system is being asked, enter X and Y and all as usual, your X is your latitude, which runs from top to bottom and then North is your longitude, which runs horizontal, okay? So here you have, since we have the degree, minutes and seconds format on the toposheet, I'm going to use that. So we say X East is, first point is 77. And then we will put, so this line is 13, 15, right? So your latitude is going to be 77 and then your longitude is on this one, the East West, so East, North. So let's do the East first. The East is going to be your latitude, which is the top down. The top down is 77, right? So we're going to put 77 and this one is 35, so 35 space. So look at the format, DD, so we put 77, 35, 35 is 35. Suppose you have zero three, you have to put zero three. Don't put just three, it needs a value, okay? So 35 and then a space and then the seconds is zero, so you can just leave it because 35 is enough. And then on the bottom, this line, your latitude line is there. We are going to put one three, because it was one three, one five, right? So space one five. So 77, 35, 15 is what we're going to do. We're going to click okay to add. So now the point has been added. It's slightly in the center, but it's okay. Okay, now you can zoom out. The second point we're going to go is the same line, 77, 40. Okay, so 77, 40 is this one. I'm going to zoom in and this is the darkest point, right? So I'm going to put it in the center and then this is a 77, 40. And then I am going to do 30, 15, right? And then we say, okay. See, I use the same coordinate system as the map. So you can change it here, but don't change it because you want the map to sit in, right? So you say, okay, now two points are done. So now four more to go. As I said, two on the top we want to put, two in the bottom. So I'm going to zoom in. So how I'm zooming in is by moving the mouse in the front. So if I move the wheel zooming in the front, you will come down and let's now come down to this location. You can see this line is 30 degrees zero and 77 degrees 30, right? So I'm going to go on this line is going to be 13, zero. So let's pick a line here with some points. So let's pick this one, 77, 30, 30, right, 32, 30. So you see this is the darkest line. So I'm going to click here. And then in the center, I'm going to put 77, 30, two, 30, and then this is 13, zero. Okay, you can just leave 13, that's fine, right? Then what you do is, if you switch the latter long, there'll be an error, okay? So be careful with that part. And I'll quickly show you if an error comes, how to change it. So the next point is 77, 30, right? So I cannot move this side. So I can use the hand tool. So to pull it, and I want to do this 77, 37, 30, and 30. So I'm zooming in, and I see a big black line, which is good enough for me. I'm going to zoom into the center, and then put 77, 37, 30, and then this is 30, zero. Okay, now I put okay. So now if you can see, we have the four points, the four points have come here. And then the DX, DY, pixels, residuals, these are errors. Once you set up the transformation, you can find if the errors are there. Okay, there are multiple linear and other things that can come. But I'll tell you which one to use. One number two is good, nearest neighbor. The target coordinate system is the same that we want. And then the output raster, you can say a name and store it. Compression method, LZW, et cetera, et cetera. But we'll come back to that after we do two more points. So I'm going to zoom in to this area because this is the area we want to map the water body. Okay, so this is a Yalahanka Kere, which is a lake. And I'm going to go on top of here and find the location. Okay, so this location is what I'm going to put. This line is what I'm going to put. How do I know what is the coordinates for this line? I will have to hold my pointer on this line, pull on this side first. And then I see that it is 730. So 70, 13, 730, right? So this line is 13 degrees, 730. And then I'll go back to that again, the lake. Make sure you don't pull up and down. That mistake you can do, I missed it here. So it's here, okay? Yeah, we are here. So now we go up to find the line on the top. So we find it to be 35. So 77, 35, right? So I'm going to come down again to that point. I've noted it in my book. Now I'll go to Yalahanka Kere. And this is the location where I want the point. Of these two, the more darker one I would say is this line. So I'm going to put it right there and call it 77, 35. Whereas this is 13, 7, and 30, right? When you click, okay. Now after you click, okay, what happens is one more point we'll take. For the method that we're going to use, we need at least six, okay? So now the one I'm going to look is at 42, 30. So why I've taken that line is somewhere here I want another point, which I don't have. And when you're zooming in, you should look at certain points that you could use for your calibration or eyeballing, right? So that would be easy for you. For example, on this, you can say, okay, this is a sheet rock. So near sheet rock, I can put a point or this one is good, right? So 916 is there, near 916. I'm going to take this point, okay? So 916 is my, visually I want to use that point. And then from here, I'm going to look in the left to see what is the, what is the latitude? So it is 13 degrees, five minutes. So I'm going to put 13 degrees, five minutes for the 916. And then on the top, it is going to be 42, 30. So it is 77, 42, 30. So let us put that value. Again, I don't know where it is. So I'm just going to see 916. There it is. I've made a note of the point. You see it's beautifully dark. So this is the center of the pixel. I'm going to go here and type 77, 37, and 30. And then down it will be 13. Oh, sorry, it was 77, 42, 32, 30. And then down it was 35, right? Yes, then you say, okay. And then you could see that these populate. So six is needed, six we have. Now we have DX, DY, residual. So now it's all 00, trying to show that the error is very, very, very small. And you can go to the cycle mark or the gear mark where you have to select which type of transformation you want. As I said, for this particular map topo sheets, polynomial two is good. It is a method of transforming and applying the lat lengths throughout. And then you have the nearest neighbor method or multiple resampling method. We'll take the nearest neighbor method. And then here is where you want the output raster to be. I'm going to call it as the D34-12 is fine. Geo-tagged, okay? So geo-tagged or geo-referenced, okay? And then, and read the errors. Don't make it too big, it won't save. And then there are a lot of compression methods. The best one known is the LCW, so let's keep it. And then the others can be a default. Load the map when it's done. You can have it. Don't save the GCP points for now. We have to see how the error is. Okay, once this part is done, you will click the play button which says it's going to run. So now the progress is going on. And beautifully you could see that still is zero because it is exponentially, very, very, very small. Once it loads, now it will load onto the map, okay? So now if you see that, the whole of India is there, but suddenly something is happening here, okay? Something has happened here. You can zoom in to see that. Beautifully, our map is now placed inside of Granataka. To make sure that it is more accurate, you can open this map in Google Earth Pro or Google Earth Engine, which we will do in the next class. But for now, we have successfully downloaded a toposheet and then found where accurately the points can be put. Those are called anchor points or GCPs. And then we also looked into how many points, six points. So you can see how the six points have been captured. And now once the points have been captured, your map is ready to be used in GIS, okay? So how this image can be used in GIS? We will see in the next class. Until then, please try multiple times. Do not switch your lat and long, east and longitude. Be careful with the naming. Your east, x, east, which is given in the georeference tool is latitude, okay? It's not x, x is, runs like this. X can also run like this. So x is your latitude and then your y is your longitude. And then you zoom into that line which intersects and then find your exact center. So if you look at this, we have put our points very carefully in the center, right? And then you can move and stuff. But again, the map has come pretty well. At least it is in Karnataka for now. Let us look at it, how good it is in the next class. Next class, we will look at it in Google Earth Pro. Google Earth Pro is a very easy software to use. It is free open source. So please download it and then use it. We don't promote any software here other than open source software. So since it's open source, it is available for public. Please try to use it if needed. But now I have shown you how to download a particular map from government of India's database and look at it in GIS interface. So initially it was an image, but now it is geo-reference. To showcase it, I'm going to open the initial image again. I'm going to go to the DSR, okay? So but before that, one more thing you need to do is right click, save or export. Export as layers, why? Because as I said in QGIS, it saves your memory by running it on the fly and putting it in the cache bin. Cache memory will be deleted once you delete this or close this program. So please save it. So I'm going to save as the same name can be used. And what type of format, geo-tiff, all these formats are there. All these have geo locations plus data, okay? So we'll keep it with geo-tiff, which is mostly the common method used. The final name you can give is, you have the same name we can use. So I'm going to copy paste this name again. Okay, so just let's do it again. I'm just going to right click to copy the name, okay? And then same name I can use. So right click, export, save as, go to the same thing, click your folder. I'm going to take the folder, I'm going to save. You say, okay, says it's already, no. You can say that geo-reference NP, NP for NP term. And then geo-tiff is being saved. All the others can be default. You can say add save file to the map, okay? You can say, okay, okay. Sometimes it doesn't store it. In this time, maybe it has stored it, which is also good. And now we have the map. So I'm going to remove this for now, okay? And then this is Karnataka again, and you have the GCP points, okay? If you close the transformation, the GCP points will go. If you want, you can save it. So it's asking if you want to save the GCP points. We can discard it because we already have finalized the model. So now I'm going to go to data again. I'm going to click here. This is the initial image. I'm going to open add. And then I'm going to also open the geo-reference JP, NP, right? So it's a GIF file. Let's open it and close. So now when I add the GIF file, it opens here. But where is this file? This is the original file, right? So you would see that if I zoom out, you can see the zoom to layer is here. And if I do this full extent, you can see India here, right? So India is here, whereas your Karnataka map, which was not geo-references here. So I'm going to close this. So if I know which one it is by right click on the click button here. So I'm going to remove it. We don't want it. Okay, and then I'm going to zoom to this layer and you can find it in India now. So in India, this layer is there. How are we going to use it for data correction? You will see in the next class. Until then, I hope you try this exercise. I'll see you in the next class. Thank you.