 Welcome to remote sensing and GIS for rural development NPTEL course. This is week four lecture two. In this week's lecture we have been looking at using QGIS and introducing GIS concepts so that we can analyze the remote sensing data. I hope all of you have downloaded the QGIS software which was shown in the last week and last class. So today we will install it just to show that how the installation process goes on and to make sure that you don't get stuck in the way of installation. So let's go ahead with the installation. I'm going to share my folder. So in your folder you would have seen this part where you have installation. You have double click this part where you could go for installation which I downloaded in the previous time. And you will come across the installation window like this. I hope it's visible. I'll just make it visible again for those that have to stop share and share. So now you will be able to see the installation web page and I am going to showcase the installation now. So in the installation you will be seeing this part. So now you could see that the installation process has started. It does take a little bit time and you will click the bottom screen. So you just say next. In the next window you will have overview of the license, the RASA plugin which is a type of data. We will cover it today. And then MRAS ID, RASA plugins from GDAV, ZIP, Oracle Instant Client, etc. So you just go through all the license agreement. We cannot read it in the class time but I recommend you to fully read it. I have read it and so I will accept the terms. The long page to make sure you read it. It's a generic statement. It's four years, five years old. So then you go next and then pick a folder in the C drive where you want the software to be installed. Normally automatically it picks the C drive. C drive is more stable. Do not put it on your desktop. Do not put it on your other drives where if any error comes it will just get lost. So C drive is normally used for installation. Create a stop shortcut if you want and create a stop menu shortcut. I click both but if you don't want you can unclick and then click next. Click install to begin. So it will start running but before that in the bottom of the screen you can see the pointer. It will be asking can you change the software? Is the software allowed to change the setting on your system? You should say accept because it is going to put software on your desktop or laptop. You will have to accept it. So in the meantime while it's installing I think we should be able to see slowly the status improving. It takes time and also I'm recording the lecture. So it will take time but let it go through. In the meantime we will look at the types of spatial data. So this is a very important topic that we need to cover for the lecture. GIS data is of two types. So you download the data from remote sensing and you collect data from observation data. There are two types of data. So this is kind of the basics of GIS. I will not get fully into the basics but just some part of the basics we will cover. So all the two types the first type is raster which is very important for this course. It is a remote sensing and GIS. So most of remote sensing if not all are raster data. So the data that you're going to use in GIS for spatial data is two types. One is raster one is vector and the first one raster is also a continuous data wherein throughout the map throughout the canvas you will have data. Not in only some pockets. So that is important to understand because you will be looking at the status. I will just check the status now again. Okay so here now it is asking me my software is asking me should I allow the software to run? I say yes. Once I say yes it starts computing, validating and installing. So let it install in the background while we discuss the course topic of the data. Okay I'll keep it here so that you can also see how it runs. So as I said there are two types of data. The raster data is more continuous. Examples are air photographs taken from reconnaissance survey planes that take data from the ground up and also scanned maps, topographic maps any image which has fully covered with data. For example if you have a sheet of paper okay so you have all this data in it right all the colors and so which means the data is full all of the pixel every pixel is covered with data. So like that you will need to look at data as across the canvas or only in some points. If it is across the canvas it is called raster data. Most remote sensing data is raster data. Okay the vector data is of discrete types okay where vector is discrete features a layer comprised of individual points either points, lines or polygons. Example, statues are points which is point on each location, roads are lines and state boundaries are polygons. We will cover each of this separately in this week's lecture. This week we will look at vector in particular in the remaining lectures raster will come next week. I will devote one week of lectures for raster because raster is more important for remote sensing data. You could see here as I mentioned most data is remote sensing data is raster data whereas your vector data is discrete features you can convert between these two data formats but it is advised that some features are lost so to keep these two separate and only change if it is needed. So let us look at this data concept in GIS in particular. We have spatial data where is the location okay so spatial and attribute data is imported in GIS where is the data taken from is called the spatial data. It gives you the specific location of the data in GIS each data has a data and information which is attributes but also the location where the data is collected and that is what makes GIS different from other softwares. So we should be able to capture that part very very carefully because we have space location data inbuilt in GIS data. So it is stored in the shape file geodatabase or similar geographic file whereas attribute data is descriptive data about what how much and when so this is the actual data let us say sensors data. So sensors data is population data it is data about how many people live how many males females let us say. If I am just giving data itself it is attributes you have males females total population attributes but we are also having a location attached to it let's say Tamil Nadu's population. So Tamil Nadu is the spatial location which is the first bullet then the second bullet is what data comes into the system. So these are specific characteristics at that location natural or human creator. These are stored in database table in a table format. So we have both spatial location now and table data associated with that location in GIS. So in GIS normally they are maintained separately in a GIS environment you will see them the data being managed separately still copying the files I will just check if we have done with the files it is still going on let it go on. And what we will do is the so we have GIS systems separately traditionally storing them in different files but now and they join them and display for analysis. So when I talk about the structure of data you will see that one GIS file will have seven to eight associated files and in one of the associated files the location of the data is stored whereas the data as table is stored in another file. When you open this file this shape file in GIS software it combines this and projects the data as one data okay so that is what the last bullet means and then we have representing data with raster and vector models. So we have here different models of data right so we have data that represents both raster and vector we will see in a real life scenario how it is done okay raster model it is area covered by grids as I said if you have the map throughout let me draw it for you yeah so you have a map and in the map or the map area every single area is having the data this is in raster format. So how is this accomplished it is divided into pixels of equal size so these are grids okay so I am not doing you know straight straight but in a normal situation each data is going to be gridded I will show you an example but I am just showing you how and each point each grid will have data in it okay so that is why it is called continuous equal sized and square cells let me clear this page okay and so what happens is you have equal sized cells attributes are accorded at each single cell where the value is going to be placed and you will have a majority of the features in the cell for example land use if in a pixel you have 70% as forest 30% as urban the pixel will be colored green which represents forest yes 30% is urban but because the pixel can take only one color one value it will take the majority color image data is a special case of raster in which the attribute is a reflectance value from the geomagnetic spectrum again as I said a satellite or a plane is taking an image remote sensing object is taking an image of the planet of the land and it is when it takes an image is the reflectance how much light is reflected and in what colors so when the reflectance happens the reflectance is monitored and captured in the camera and converted into digital image in the digital image coloring is based on how many colors the camera can capture so cells in image data often call pixels for picture elements in a vector model the fundamental concept of vector GIS is that all geographic features in the real world can be represented either as points or data or dots nodes trees poles fire plugs airport cities like they are as points okay so imagine we have I will just check if the software has been updated it is still running so let it run you can see it here yes so what we will do is we will have more so we have more points or dots representing trees poles figures in the vector model in the meantime I have noted that the software has downloaded successfully let us quickly look at the software for which I will open so now if you could see my other screen just open the other screen so that you can see so if you could see the screen this is where we downloaded the software and if you click your start button on the top you could see that your GIS has been set up basically your GIS 3.2 to 0.14 and what location it has come has been there so I'm going to open the 3.2 to 1 4 which is the long stable version and then your 3.28 is your newest version with multiple features the feature which one is there remember that you should remove your previous ones I have my older version also but that's the beauty of QGIS it lets you run multiple versions because some people like the older version they have the older version and the newer version also in the same system so once you click your QGIS this thing opens up which says that welcome to the long-term release the version is given and the name each a model is given a name bio tovizia or I don't know how to pronounce that last dot is that a so these names are taken and given by QGIS and it starts so always you open a new product and then it says some plug-in has crashed you know the the warning but now your QGIS has been set up so in the next course lectures I think week 5 6 we will have hands-on on how to use this so I'll use this video for that but I just wanted to show you how to install it into your system and then take it up so let's get back to the lecture notes that we were presenting okay yes so we stopped here where we said points or dots nodes trees fire plugs airports and cities where you have a vector model a location is taken for these as a point and for the point data has been stored the data can be anything but examples are trees poles fire fire plugs airports and cities and the that is one type of vector data the other type of vector data is lines where you have a combination of points joined together as a line and the location of the line is also split in different points we will not see that part but we will definitely see the line in your image and a perfectly placed line in the geolocation it's not like randomly placed line it has a location and it will be placed likewise okay and then we have areas so a combination of lines become a polylines and then polylines become polygons if they are connected and closed so areas are polygons where it is again starting with points it converts to lines lines convert to polygons and polygon examples are land parcels cities counties forest prog type so these boundaries can also be put on a GIS map and with the location and with the data so the last part we would like to see here is a video has gone up let me the last part we would like to see here is because representation depends on shape GIS refers to files containing vector data as shape files since it has a shape point is a shape line is a shape and a polygon is a shape we call this as point files let's look at a real world so that you can understand how the system works in a real world of concept the top part is an is a real world where you have a river flowing so the line here is a river and then you have houses forest trees and then houses on both sides of the river okay so now i'm going to convert it into the two types of data we have in GIS raster representation and vector representation let's do the raster first because we have discussed that as i said the image is first discretized into grids equal size square grid grids the size depends on the spatial resolution of the camera so here you could see that it it has been divided into grids each grid gets a unique number this is 0011121314 etc so we don't care about the labeling scheme but for now rows columns are given nine rows nine columns and then the grids are being established so each grid is of equal size and inside the grid there is only one value for data you cannot have houses and rivers in the same box it's either house river or a blank or a tree so there's only four data types there is blank this is land or let's say land and then there's a house there is a tree or a river only four types of data that can be represented in this image so here in the raster representation you see the blank which represents the land you see the h which is the house and then the river going as a connected grid as r and then you have t for trees you could see that it is not like three trees are here four grids are colored no it's the space the space occupied by three trees is represented by four grids here and then one one there and then a house so all these are on the location of where the data is being collected so this is the raster representation of the real world wherein we have the real world painted as a whole canvas in the raster and but in the vector only the location where exactly the house is here also there's a location but the whole grid is colored as house whereas here the location exact location only a point is kept for houses the trees are combined as a forest as a polygon so all this area is taken as a polygon and then the river is created as a line so here you see the river going through as a line and the data points line and polygon has been covered so this is very important for understanding the conceptualization of a real world into raster and vector and then we move on to one more example in the next example I will also show you how the 3d effects are going to be captured or not captured in a raster so here going from bottom up we have the real world you could see that the real world has some imperfections and undulating surface it's not a flat surface but GIS is about converting this onto a 2d frame your computer screen is a 2d frame so you're going to put this image on a 2d map and then each data point can be converted for the 3d image so in a real world you have a land mass some water body a lake and then streams that get into the lake a forest and a marshy land so this is a real world imagine that you are having a real world where you have different types of attributes or data in a vector world how it looks like is you are converting that into a 2d surface first and not all parts of the surface needs data but still the data is represented as grassland which is the brown and then this part is the marshy land where some green is there so light green is given forest comes on the top north part of your study area so that part is a polygon also at forest is being mapped and then the streams are lines that drain into the another water body polygon which is the lake so you have the lake the forest and the grassland the marshland as a polygon and then you have lines also in this vector representation you do not have points but that is also okay because point can be used for marshes or forest but in it is not necessary that all vector data should be represented in your image it can be purely in a map it can be purely points it can be purely lines or polygons or a mixture or combination of either you do not have to have specific all of them or a partners and stuff any number of data as per needed can be used now come to the raster world in the raster world it is not as straightforward you will have to convert the entire 2d canvas into grids and all each grid has to have a data there is no part in the grid which says no data so you could see that all of them are colored brown and on top of it there is marshland which is colored yellowish green and then you have the blue which is the river and the water bodies and then you have dark green for forest the green color is is looking at the forest which is being reflected back as green so what do you see here very carefully is the grid size dominates for these kind of analysis for raster because the thickness of the river can be small maybe small but if the grid size is big the entire grid is going to be colored as blue so you could see here that the majority is blue so all of the grids are showing blue but this line might be smaller this line might be smaller than the top line and that cannot be differentiated in a raster because the raster will still think that majority of my grid is blue so I will color it blue so this is the only issue and it is not as clean so you could see that the grid is just connected one one one one grid to the lake so it is not as as natural as a river flows so it doesn't flow as blocks correct so in raster you will see that the data flows as blocks whereas in a vector it is moved as lines and stuff however it is not continuous as raster so that is where there are benefits of using either data set so one data set is not better than the other data type it depends on your use how you're going to use the data and what you're going to use the data for so moving on the differences here we see are the raster data has cells and pixels the grids as I said is called cells or pixel or you can interchangeably use them across your understanding grids is equal to cells and pixels and each of them have a property that across the grid across the canvas across the image the grid size is the same and it is a square grid you cannot have a rectangle on the top and then square on the bottom also the square should be of same size the size again is divided and estimated by the cameras and the data acquisition system so we will not get into that but understand that whenever you download a raster data you should understand what is a grid size which is a spatial resolution and the grid camera or whatever has been used to form the grids in the next part which is vector you have features so data is divided as cells pixels grids in the raster whereas data is divided as features in the vector type so here you could see marsh lake forest those are names but those are also features under which data are stored as attributes so you have marsh so in the marsh you can have type of marsh size of marsh length breadth all these things can be added as data the major item is your feature and within the feature you have attributes so it's like your column name could be your feature class whereas your rows in a table could be your attributes the grassland is a type of feature the lake is a type of feature the entire table can be called as land use land cover classification and that is being used to represent the real world scenario again the real world is very complex but you're trying your best to capture it in a raster or a vector format please understand that because there's a lot of grids and more representation continuous data raster is always bigger in size it consumes more memory storage compared to vector vector you can show a lake as a point a line or a polygon only part of the map area is going to be used whereas for raster even though a small lake is used the entire image has to be classified you cannot just classify the lake grid area you will have to classify the entire area so these are the pros and cons of using the data types there are two data types in GIS raster and vector and raster is where a continuous data set is created by discretizing the image forming grids whereas in vector the important features that are needed to be mapped are identified they are converted to point line or polygon and then they are input into the system with this I will stop today's lecture I will see you in the next lecture where we will specifically look into the vector data set and how it can be used for rural development whereas in the next week we will look at the raster data set and how it is used for rural development again those who would like to learn more on these type of data sets and GIS please take the recommended NPTEL courses and or any courses that talk about GIS data types it is the same it is not different because different people teach it it's a very basic simple entity we do not teach it here because then the load on GIS will be more rather than rural development the understanding is you will learn this or you had already learned GIS data types before you came here but it's fine if you don't please learn it in the side we will be using that in the rural development concepts thank you