 Hello and welcome to today's lecture. So, right now we are in module 2 of the course on Microwave Remote Sensing and Hydrology. And before I begin the lecture, let me give you a quick sneak peek into what is included as part of this module. So, it is titled Imaging Radars and How Do I Interpret? So, here we shall be learning about what is synthetic aperture radar abbreviated as SAR and how are these images formed? We shall understand SAR as a complex number and learn about the different types of SAR imagery. In addition, we shall also be learning about key terminologies like azimuth resolution, ground range resolution, a backscatter, sigma naught and so on. And slowly, we shall introduce ourselves into the image defects, the geometric and radiometric distortion and we shall learn about what is peckle, the different data formats in which SAR data is being made available and what are ratio images. And slowly towards the end, the concept of texture, contrast and polarimetry shall be discussed. And remember parallel to the lectures, we shall also learn how to work with SAR imagery in SNAP and Python through the upcoming tutorials. Alright, so with this background, let us move into our first lecture of module 2. Let me begin today's lecture with a question. You know, when we mention about SAR images or for that matter radar images captured by satellite, where do these satellites exist? Now for that, we need to go and view the earth from a higher altitude. So, what you see here is the Jim Khanna ground of IIT Bombay. So, let me move a bit farther and farther away into the exosphere and let us assume that these satellites are say about 600 kilometers from the earth surface. So, assume we have a satellite which is looking towards the earth. So, the region that is illuminated in light blue color that is known as the footprint of the satellite that is at an instant of time, the area on the earth surface that is viewed by the satellite is known as footprint. Now again, you know, let me start with the question as in how do satellites detect microwaves? Since we are talking specifically about microwave remote sensing. Remember the example I gave you in one of the earlier lectures about radar guns and then I mentioned that radars send out microwaves which then hit the target, come back, they are received back at the receiver and the time of travel is used to compute the range that is the distance from the target on the earth surface to the satellite. So, let us try to understand this clearly comparing it with echolocation of bats. Now bats send ultrasound pulses out and then they get the echoes back to create an image of what it is seeing. Similar to radars that send out microwaves just like the microwave used in your kitchen oven, you know, it sends out microwaves and then it gets the echoes back from the ground to create an image. Now again, why am I comparing bats and radars? Because both of them can see in the dark. So remember when I mentioned in one of the previous lectures that microwaves have the ability to penetrate through clouds, which means they are not blocked by clouds, they can see in the dark as well as through bad weather. So, it provides all day, all night weather sensing, pretty powerful. So, now I mentioned that radar sends pulses and these pulses hit the target. So, by target here I mean the earth surface and then the echoes that return from the earth surface travel all the way through the atmosphere and they get captured by the satellite. So, here as I mentioned earlier you see the footprint and then whatever is highlighted in red is the echo that is getting scattered and then which is pointing, returning to the satellite and of course you have some amount that is lost, you know, that are getting scattered in different directions which are again indicated by small elliptical drawings. So, you have the footprint, the radar sends the pulse, the pulse hits the target, some amount is scattered in different directions and some amount of this is scattered and then it is returning back to the satellite. So, for an example, let us try to understand how a single pulse is traveling. Now shown here is the illuminated area from the instrument. So, now I am not showing the instrument, I am showing the illuminated area and now we can see that the pulse sent from the satellite is traveling towards the illuminated area and then it hits the target and the resulting echoes are getting scattered. As I mentioned before, some of the echoes go back to the radar and they are the ones that shall be registered to create the image. Now, once again what the satellite would observe would be these few echoes that are getting scattered back. Now, again, you know, you see two echoes going back. So, I am confused because how does the satellite differentiate between two echoes that come from two different targets because we need to understand what target has caused what echo and then how bright it appears on the radar image which means there should be some mechanism by means of which we need to separate these echoes from each other. So, if we can estimate the exact time at which each echo was received back by the satellite, we can calculate the exact distance of the target from the radar, is not it? And by some means we need to differentiate the echoes from one another. So, for this we need a solution and to explain the solution, let me take the help of sound waves. Assume I am playing you three keys, I am taking the example of sound waves and if you are a music lover I am playing you three keys that is three piano notes in E4, A4 and G4 frequency which is mentioned here. So, let me play you the first audio. You hear the three piano notes separately, is not it? Again why are we doing this? Because we are trying to understand the mechanism by which satellite, radar satellite gets to differentiate between the different echoes that are returning back after getting scattered from different targets on the earth surface. We are trying to understand it through a simple example wherein first I played you three piano notes. Now, let us introduce an echo. You hear the same notes but you are able to identify or differentiate when the echo is received back, is not it? Because every time an echo is received the pattern of notes is repeating itself from the beginning and I am able to identify it clearly. Which means if I replicate this somehow in imaging radars, we shall be able to identify and differentiate between individual echoes. So, what did we do? We tried listening to three piano notes separately without echo and then the same three piano notes again with echo and because you are familiar with the pattern of notes, it is easy for you to differentiate when the echo has occurred correctly because you are aware of these notes. In a similar way, if an imaging radar has some mechanism by means of which this can be replicated, it is easy for us to differentiate between echoes as well as differentiate between the different targets that has resulted into this echo. So, let us move forward. At this point, it is worthwhile to remember that when the pulse is incident on the near part of the footprint, that shall send the echo first. And as it moves through the footprint, more and more echoes will be sent out. Now, there is still one more problem because we do not know where the echo has exactly come from because there are certain regions which are denoted by the blue lines within the footprint which are equidistant to the radar, which are at equal distance to the radar. So, which means if there is an echo, it can be from any point on the line but we do not know where exactly the echo has come from. So, let me try to explain it with a simple example. So, what I have done is I have divided the footprint into two. One is the shaded hatched region here, the front portion and the backside is left as such and there is a center line separating the two regions of the footprint. This is the center of the beam. Assume this is the center of the beam. So, remember the footprint is not stationary as the satellite is moving, the area illuminated on the ground is also moving. So, all of these that is towards the right side, they are moving towards the satellite with the satellite and all of these that is towards the bottom, they are moving away from the satellite as the radar is looking sideways and as footprint is the area that is illuminated on the ground that is the area that the radar sees as it is moving. So, here let me try to introduce the Doppler effect and we gave the example of an ambulance or any vehicle with a siren. When it approaches you, you observe a change in pitch and even if the vehicle with the siren is stationary and you are the person moving, still we observe a change in pitch that is relative. So, let me come back to the radar. So, given the pitch of an echo, we can estimate where exactly the echo has come from. With all this background, the next question is how bright do these echoes have to be in an image? Which means, these echoes how do they appear in the image? How bright do they register in an image? So, at this point, image we can define it in the form of a matrix of numbers, they are made up of different pixels, image made up of different pixels where by pixel I mean the smallest unit of an image and generally we consider image as a matrix of numbers. So, in simplified terms, small objects shall give rise to smaller echoes and larger objects shall give rise to larger echoes and these eventually get recorded in an image something like this, with bright and dark pixels. Larger echo, medium echo, smaller echo. So, if we zoom out slowly let us try zooming out what you see here is the IIT Bombay campus and then you can see the Powai Lake and let us zoom out a bit further, we see the Mumbai Metropolitan Paths of Maharashtra and what you see here is a sample of synthetic aperture radar image from Alo's Pulsar that is captured in the L band. So, now we know how a synthetic aperture radar or SAR imagery is going to look like. Remember we have still not introduced the technical terminology, we have been trying to understand it in simpler terms. So, what we will do is let us try to talk a little bit technical now. So, we started with the operation of a radar and that is when I mentioned that a radar system is able to effectively measure distances by transmitting large power microwave pulses to illuminate a target or a surface and by making extremely precise echo delay timings. So, the two way distance between the antenna and the target which is known as the range, the distance is known as the range, it can be computed by using the simple relation c tau by 2 where c is the speed of light, tau is the duration of pulse and why are we dividing it by 2 because the two way travel is considered, travel from satellite to the target and from target back to the satellite. So, then let us come back to our original question you know how to detect or collect and then quantify microwaves. So, remember I am making a distinction between collection and then quantification because our interest is to estimate the average intensity or the energy of radiation at microwave frequency. Of course, it may be of interest to measure other properties of the electromagnetic radiation like say polarization or the phase. But remember I am making distinction between two terms here how to collect the data by data I mean microwaves right now and second is how to quantify it. So, let me introduce antennas here which collects the incoming radiation. So, till now in the initial parts of the lecture we were just referring to satellite sending pulses and satellites receiving pulses. Now, let us be more specific and I am going to mention that antenna collects the incoming radiation antenna on board a satellite and receiver amplifies and detects this collected radiation ok. Sometimes a third element is also essential which is specific to measurement technique which is data handling system which is normally used for digitizing, formatting, calibration, housekeeping of data etc. So, since we are talking about imaging radars. So, imaging radars generally they tend to have huge data rates that the bulk of processing has to be carried out by a separate system within the ground segment. We will discuss about that shortly in upcoming lectures. But just to be clear in remote sensing literature the visual metaphors dominate you know it is the visual metaphors that exist widely because always in remote sensing the eyes and the camera lens are considered as examples of remote sensing instruments which capture information about a target without coming into physical contact with the target. Unfortunately, you know our familiarity with remote sensing in the optical region or visible region is not going to be helpful here because we are discussing about microwave remote sensing. So, if I have a background on remote sensing I will be refraining from understanding microwave images using the example of eyes or camera lens. So, here I would insist that you compare microwave systems analogous to your ear you know analogous to your ear. So, because we as humans how do we detect sound when you are say in a dark room which is pitch dark. So, as soon as you hear a sound you are going to automatically turn your head and your ears to the direction of sound isn't it? And this motion of your head points to the direction of highest sensitivity of ear towards the sound source isn't it? Let me play a video for clarity so that you assume ears analogous to a radar antenna which has to point towards the direction of highest sensitivity towards the source of what of microwaves. Now at this point let me again mention that the transmitting and receiving antennas can be different. So, transmitter and receiver the transmitting and receiving antennas can be different but then it is usually more efficient for the same antenna to both transmit a narrow beam as well as receive a return signal. And microwave systems as mentioned here will collect energy from a comparatively wide range of directions even though it may be more sensitive, most sensitive to a small range of angles. We will follow this up with the relevant derivations but for now I want you to understand that microwave systems they compare they collect energy from a wide range of directions even though it is most sensitive to signals within a small range of angles. So, in camera an aperture opens to let light in isn't it? So, similar to that in microwave remote sensing an antenna on a spacecraft or an aircraft it can be considered as an aperture synthetic aperture radar okay aperture. So, an antenna on board a spacecraft or an aircraft can be considered as an aperture and re-iterating and the use of decibels in microwaves is very common okay just a fact for us is our ear sensitivity to loudness has a logarithmic relationship with power okay. So, till now we have made a distinction between antenna and a receiver and we mentioned that it is usually efficient for both the antenna to transmit as well as receive the signals. Now let us try to understand what is synthetic aperture radar SAR. So, generally larger or longer the antenna more is the information that can be obtained about a target okay but in real world antennas in space are not that large or long. So, spacecraft's forward motion is used along with signal processing techniques to simulate the existence of a long antenna large antenna. So, what are we doing? An arbitrarily long antenna is synthesized by recording the echo or the signal as the satellite moves forward along its orbital track and the results are combined with appropriate techniques which we shall discuss later but for now I want you to understand how synthetic aperture radar functions. Synthetic what does synthetic mean? It means something that is not existing okay you are artificially creating. So, here by synthetic aperture radar I mean an arbitrarily long antenna is getting synthesized because with the movement of the satellite along the orbital track signals are being collected what you see here are the footprints it is also combined with appropriate techniques to generate synthetic aperture radar imagery okay. So, with this background let us try to discuss in detail about SAR images. In the first module we discussed about electromagnetic waves what are amplitude, what is wavelength, what is phase, what is frequency. So, whenever an electromagnetic wave is getting scattered from a position say a position x, y, z on the surface the physical properties of the terrain cause change in both the phase as well as amplitude of the wave okay. Let me rehydrate an electromagnetic wave when it strikes a target on the surface of the earth the physical properties of the terrain are going to change the amplitude of the wave as well as the phase of the wave and what a synthetic aperture radar measures is this number pair that is a cos phi a sin phi where a is the amplitude and phi is the phase because of which SAR data are also known as complex images okay. In the middle of the lecture I mentioned that image as a matrix of numbers now let us go a bit further and say that SAR data that is synthetic aperture radar data are also known as complex images. Now revisiting SAR as a complex number we already discussed what the bright and dark pixels mean that is the strength of echo received and if I zoom into one single pixel it is going to be a complex number with real part as well as imaginary part okay just for you to visualize what is happening because the phase as well as amplitude is changing whenever electromagnetic wave is striking the target. Now there are different types of SAR imagery okay because from the complex image a variety of other products can be formed for example images of the real part images of the imaginary part amplitude images phase images intensity images log intensity images and so on. So using the complex image of SAR using that we can also create a variety of other products. Now here the use of intensity is synonymous with power or energy. So shown here is the amplitude image again from LO's Pulsar data by now I assume that you know what is V and what is H vertical and horizontal polarization and shown towards your right side is an intensity image in VH polarization vertical horizontal okay and this is again of parts of Maharashtra. So typically the real and imaginary images they show some structure but then they appear extremely noisy you can see the noise which is pretty evident in both these images isn't it and usually the face image is noise like and does not show any structure while the amplitude and intensity and log images though they are noisy they are relatively easier to interpret. So which means out of all the types of SAR imagery the amplitude and log images are preferred usually preferred since the large dynamic range of intensity image can reduce the perception of detail okay. So to summarize in this lecture we started our class with an understanding about imaging radars the basic operation of a radar and in simpler terminologies we understood how a synthetic aperture radar images formed we understood what is an aperture and then put together what is synthetic aperture radar how do SAR images look like and what are the different types of SAR imagery and we have just started understanding SAR images as complex numbers. I hope you found this lecture informative and I will see you in the next class. Thank you.