Complete Drone Surveying Crash Course: From Setup to High-Value Deliverables

In this 1-hour crash course, I'm going to show you everything you need to know about drone surveying. We'll cover

everything, including picking the right drone for your projects, how to establish high accuracy survey control,

planning and executing your drone missions, processing your data using specialized photoggramometry software,

and the proper way to assess the accuracy of your data and create high accuracy survey grade deliverables that

pay you thousands of dollars and help you gain more clients and build your business. My name is Rammy Tamimi. I'm a

professional surveyor and I've been a UAS pilot since 2017. I won the younger geospatial professional award at Geo

Week in 2022 and I've been making surveying and geospatial technology videos online for over 5 years. I also

have a premium survey community called the survey school where professional surveyors, drone pilots, business

owners, civil engineers, and all sorts of disciplines come together and learn about the fundamental principles of

surveying and how they can implement that to geospatial technology to effectively grow their surveying

knowledge. And this drone survey course is a taste of what you will get inside of the survey school. Now, before you

get started, make sure you're in a quiet environment where you can focus and take notes on everything that I'll go over.

And for those of you who make it till the end of the video, I'm going to reveal to you how you can get a free

license for Pix 4Datic. That's the specialized photoggramometry software that I'm using in this course, as well

as a free IMLid Reach GNSS receiver. That's the GPS unit that I'm also going to be using in the course. All right,

sit back, get ready. Let's get started. Now, I'm assuming many of you have gotten your FAA Part 107. Just like any

commercial drone operation, in order to do drone surveying, you have to have your Part 107 license. Even if you are a

surveyor, the FAA makes no exceptions. So, you have to have your 107 in order to do drone surveying. So, if a survey

company doesn't have a drone pilot, you may be a favorable contractor to them if you understand how to do drone

surveying. Now, there are many drones that can be used for surveying. If you're just starting out and you're

looking for a budget friendly drone, then I would recommend the Phantom 4 or the Mavic 2. These drones are fairly

inexpensive and you can buy them on Amazon. Now, while these drones do have a decent camera, they do have a linear

rolling shutter or an electronic shutter. I'll explain in a minute what that is, but this does affect the

quality of the imagery. Now, this doesn't mean you can't use these drones. You'll just have to fly a little bit

slower and only fly smaller sites. For someone that's just starting out and doesn't have a lot of money, if you've

gotten your 107, this could be a decent option to get started. Now, when it comes to an intermediate drone, I would

recommend the Mavic 3 Enterprise. And this is the drone right here. This is what I use on a daily basis. And I

recommend this drone to anyone that's starting out and is willing to spend a little bit more money. You can pick one

of these up for less than $10,000. And as you'll soon learn, the sensors that come with this drone make it a very

desirable option for drone surveying. Now, this camera sensor right here is a 20 megapixel mechanical shutter camera.

And the main difference between a mechanical shutter camera and a linear rolling shutter camera is the way that

the shutter operates in order to capture an image. A mechanical global shutter camera blacks out the entire scene

before capturing the image, making the entire image captured at the same time. Whereas a rolling linear shutter rolls

the shutter while the image is taken. And so if your payload is moving like a drone while you're capturing the image,

then you may experience some motion blurriness and this affects the quality and accuracy of your data when you're

doing drone surveying. This may work really well for someone who's just taking random real estate pictures. But

if you're doing high-end work like drone surveying, you're going to have to get a mechanical shutter camera. It's just the

best way to do it. The other great thing about the Mavic 3 Enterprise is this sensor right here. This is an RTK

antenna providing this drone with centimeter level accuracy when collecting imagery. Now, every image you

get out of a drone has a GPS geo tag attached to that image. The accuracy generally ranges between 3 and 5 m, but

with RTK, you can achieve centimeter level accuracy. This allows us to do drone surveying by achieving a higher

level of accuracy when collecting the data in comparison to an entry-level drone. Now, real time kinematic or RTK

corrections is a method used to achieve realtime corrections to the drone's GPS position. So, having this antenna allows

us to do this, which is just one more reason why the Mavic 3 Enterprise is a great drone for survey mapping. Some of

the more advanced level drones include the DJI Matrice 350 RTK and the Wingra 1 Gen 2. Now, these drones are classified

as enterprise drones and generally run between $30 and $100,000. The big advantage with these types of drones is

that you can remove the payload. So, while the camera on the Mavic 3 Enterprise is attached and it cannot be

removed, on the Matrice 350 and the Wingro 1, you can swap and change the sensor that's on the drone, giving you

the ability to use 42 and 60 megapixel cameras, LAR sensors, multisspectral imaging, as well as thermal sensing.

Now, let's talk about the science of photoggramometry and how it has become a gamecher in the surveying industry. Now,

photoggramometry utilizes imagery in order to create a 3D model by utilizing the focal point of several images and

reconstructing the scene in which we are collecting data. That means while our drone is flying in the air and just

taking pictures for us, it's important that we have a clear and visible image every single time so that the

photoggramometry software that we process all of our images into can generate an accurate 3D model. In

addition to creating the 3D model, photoggramometry ties georreerencing into the image. Georreerencing relates

the model somewhere on Earth. So while we are collecting data, it's important to use that GPS information to know

where we are on Earth in order to georreerence the data after we process it. And so now I've mentioned GPS

several times, but you may have also heard of GNSS. And so you might be asking yourself, what's the difference

between GPS and GNSS? GPS stands for global positioning system. Essentially, this is the American satellite

constellation. So all of the American satellites that are orbiting around Earth fall under GPS. Now GPS is a part

of GNSS. GNSS stands for the Global Navigation Satellite Systems. GNSS is a collection of satellite systems from all

across the world. So regardless of which country that satellite comes from, it is a part of GNSS and that includes

American GPS. Some other satellite systems include the Russian GLONAS system. Um I am not even going to try to

pronounce the Russian name. The European Galileo system as well as the Chinese BYU system. These four countries make up

the largest satellite constellations inside of GNSS. So GNSS receivers like this one or the one that's built into

the drone use all of these satellites to establish their position here on Earth. So how does this GNSS receiver achieve

centimeter level accuracy? Through the use of RTK. So let me break down what RTK is. In order to get RTK corrections,

you are going to need three segments. The first is the space segment. These are your satellites that are up in the

sky. You must have a clear visible sight to these satellites from your receivers. The second is the control segment. This

is a static point here on Earth that observes the satellites up in the sky over a known point. Now since you know

where the base station is located, all of the observations it gets from the satellites will have error and these

errors are then controlled by the base station and sent over to correct the position of the user segment. And so the

user segment can be a GNSS receiver like this one or a drone like this one that is moving constantly collecting data and

is connected to a base station receiving corrections for their position providing us with centimeter level accuracy. Now

it's important to note that both the base station and the rover need to have clear visibility to the satellite in

order to have a fixed solution. So for your RTK corrections to be fixed, you need ample satellite visibility and a

connection between the base and rover. If you don't have satellite visibility, but you do have a connection between

base and rover, then you have what we call a float solution. A float solution generally has about 1 m of error, but is

definitely not survey grade accuracy. And if you were to have ample satellites but no connection between the base and

the rover, then you would have a single solution. And these are the typical GPS accuracies that you get with your phone

or any single solution device about 3 to 5 m. And that's why we only collect data using a fixed solution RTK, giving us

centimeter level accuracy for our position. So, now that you have your drone picked out and you understand how

GNSS receivers work, giving us centimeter level accuracy, the next thing we're going to cover is ground

control points. Ground control points or GCPs are used to aid us in increasing the accuracy of our drone surveys. By

using targets like this one with high contrasting colors, we are going to be able to measure on the ground different

positions that will give us coordinates for ground control points that we can geo reference to our data. Now, we'll go

over georreerencing in the software at the end of this course, but basically what we're doing is identifying the

center of this target in our photoggramometry software. So each image will show this target from an aerial

perspective and we will identify the survey coordinates of each of these targets by selecting the pixel of each

image that corresponds with the center of the target. This along with high accuracy geo tags for our imagery will

give us the maximum level of accuracy necessary for drone surveying. Now, in addition to ground control points, you

probably have heard of a term known as checkpoints. The main difference between ground control points and checkpoints is

that ground control points influence the accuracy of the data set. So, I include ground control points to help improve my

accuracy. However, checkpoints are used to validate the accuracy of the data. Therefore, we don't actually use

checkpoints to influence our data. We just simply georreerence their positions and then compare the results between our

3D model and the coordinates that we measured with the GNSS receiver. This will help us establish the accuracy of

our data. And again, we'll go through all of this in the software later on in this course. Now, we've been talking a

lot about GNSS receivers. And the one that I'm holding here is the Imlid Reach RS3. Now, I recommend IMLED to anyone

that's just getting started in surveying because it's fairly inexpensive. It provides high accuracy data and their

interface is very userfriendly. This one right here that I'm using as a base station is the Imlid Reach RS2 and each

of these sensors is under $3,000, which is very affordable compared to some of the more high-end survey GNSS receivers.

Now, advanced level surveyors are probably using brands like Leica Geo Systems, but if you're just starting

out, ID is completely fine and a great brand for you to start if you're just getting into drone surveying. Now, the

way in which you collect data using a GNSS receiver requires you to hold the rod plum so that the bubble is in the

center. But the beautiful thing about the IMLid Reach RS3 is that it has a builtin inertial measurement unit or

IMU. An IMU allows you to hold the rod in any way you want and the very tip of the rod is calculated. So even if I'm

holding a point like this, that imu is going to be able to figure out exactly where the bottom of the pole is,

essentially correcting the position had I held the rod like this. And so this makes it very user friendly for anyone

just starting out to ensure that you're collecting data accurately. And so now you might be telling me, you know, Ramy,

that's great that this GNSS receiver costs under $3,000, but I need to buy two of them. Now we're at $6,000. Well,

lucky for you, I'm going to show you how you can achieve RTK corrections, centimeter level accuracy without the

need of a second receiver. Allow me to introduce you to transport of RTCM via internet protocol. or ENTRIP. Entrip is

a network solution that will provide any rover you have, including your drone, RTK corrections by connecting to

something known as the virtual reference system or VRS or a continuously operating reference station or a corors

base station. The main difference between VRS and corors is that VRS utilizes several base stations. So while

you're flying, you are connected to the internet and getting virtual corrections without having to set up your own base

station. Whereas the corors network is actually physical base stations set up by municipalities or private entities

that are broadcasting their position over the internet. And using entrip, you can connect to a corors network to

receive corrections from this corors station. So, now that we've gone over all of the theoretical concepts that you

need to understand, as well as the hardware we're going to use for our drone survey, let's start by

establishing ground control points using our IMLID Reach RS3. This is the Google Earth image of the park that we're going

to be surveying. And in order to place our ground control points correctly, we are going to set up one point at every

corner and then one point in the middle. So, I'm going to take my IMLid Reach RS3 and five of these targets. And we're

going to start by setting the first one in the northeast corner of the site. All right. So, we're going to go ahead and

establish the first two ground control points. Okay. And I think I'm going to set the

first point here. Lay it out like this. All right. There we go. So before I measure this point, I'm going to set up

our GNSS receiver and I'm going to be using the IMLID flow app. It's free to download in the app store and you don't

need to subscribe to Survey Plus to collect data. Anyone can collect data for free. I'm going to select IMLid Flow

and there we go. Our receiver has come up. Now I can connect in two different ways. I can connect either through the

Wi-Fi hotspot or the way that I prefer is the Bluetooth settings. So I'm going to connect through Bluetooth. There we

go. Now we're connected. So if I look at the top here, it says waiting for RTK and on the right it says single. Now

this right here, so these are the satellites that it's observing and based off of the red, right? And based off of

red, yellow or green shows us the strength of the satellite. So now I want to get RTK corrections. So I need to

input what kind of corrections I have. So I'm going to select correction input. And I have a few different options here.

If I was going to use the base station, the one that I had set up earlier, then I would select the Laura radio. But

since we're going to be using Entrip, I'm going to select Entrip over Bluetooth. Now, here I have several

different states that I connect to. So, I'm in the state of Michigan. So, I'm going to connect to the Michigan Corores

network. So, I'll select Michigan. Okay. And we're going to wait for corrections. There we go. We're receiving

corrections. And if I go back to the receiver and select the status option. And there we go. If we look here, it

says the solution is fixed. So, we have RTK corrections to our GNSS receiver. Now, in Immlid Flow, I'm going to start

a new job. So, I'm going to come down to survey. I'm going to create a new project. And I'll just call this drone

survey author. I can just put my name. And what's important is this, the coordinate system. Depending on the

projection for your project will dictate which type of coordinate system you'll pick. So, in the state of Michigan, I'm

going to be using the North American datam of 1983. And I am in the southern part of Michigan. So, I will select

Michigan South. And then depending on the state or country you're in, you'll select either international foot, US

survey foot, or if you're outside of the United States, you're probably going to pick meters. In the vertical denim, I'm

using NAVD88 along with Goid 18, which is the latest geoid correction for orthometric

heights. All right, everything else looks fine. I'm going to hit save. And there we go. I'm in my project. So now

I'm going to make sure that my pole is completely extended. Come over here to point number one and I'm going to add a

new point. All right, this looks good. And we're going to hit measure and we're going to measure this point. I like to

take 10 shots and then average them. That way I get the best results possible.

Okay, perfect. So now we've measured this first point. Let's go ahead and set our second ground control point. Okay,

so I'm going to set the second point here. And we'll hold the center. And this is

point number two. And again, I don't have to hold this completely still because the IMU will correct any of the

errors that I have if I'm swaying and I'm not perfectly level. Let's head back inside of the park where we're going to

set point number three. Now, when setting this third point, I want to make sure I set it somewhere that's open and

is not obstructed by trees. So, somewhere here in the middle will probably be just fine. Okay. I think I'm

going to set it over here. Okay. And here we go. I'm going to now measure point number three. Looking

good. Few more seconds. Perfect. All right, let's do number four. All right, we're going to set that

point right over here. I think this is a good spot. Should have good coverage here.

And coming in here, this is going to be point number four. All right, perfect. Looks good. One last ground

control point. Point number five. Here we go. This is point number five. Busy road

here. All right, perfect. There we go. And now we've set all five of our ground control

points. Now, in addition to setting the five ground control points, I'm also going to set two checkpoints. The first

one I'm going to set right here. I'm just going to change our code here to checkpoint. And this will be point

number six. Five more seconds. Okay, good. And the second one is going to be over here.

And this is going to be point number seven. All right, perfect. All right, so now we

are done collecting ground control points and checkpoints. And now it's time to begin preparing to start our

drone survey. So before we start, I'm going to be setting up our base station so that we can perform PPK corrections

to our drone data. Now PPK stands for post-processing kinematic and it is an alternative correction method to RTK.

Now since RTK is dependent on a connection between the base and the rover at times that connection is

compromised. This compromise can lead to a single solution observation. Therefore decreasing our accuracy and affecting

the overall quality of our drone survey. And so to eliminate that problem, we use PPK corrections to aid us in maintaining

our accuracy. Now, the process is pretty simple. You take your base station like this one and you set it up over a known

point. I'm going to be using this monument that was set by the National Geodetic Survey or NGS. And so NGS

posted these coordinates for the monument. And so I can trust that this is an accurate point that I can

reference my base station to. And so by setting up our base station, this becomes our control segment which will

allow us to correct our data. But the difference is you do it in the office and not in real time. Hence the post in

PPK after you do the data collection. Now the beautiful thing about PPK is that it doesn't require an internet

connection. If RTK is unavailable, as long as you have one known point, even if the base and rover are disconnected,

you can do PPK corrections on that one point without an internet connection, and you'll still have centimeter level

accuracy. That's what makes PPK a great option to validate our accuracy, even if we're doing RTK. Now, IMLD offers us a

free software to use called Imlid Studio. And again, once we go inside the office, I'm going to show you how you

can do your PPK corrections so that you can maximize your accuracy when you're doing your drone surveys. So, this is

the NGS survey monument that I'm going to be setting our base station over. Going to start by putting the legs right

here. And so, using this tri brack, I'm going to level it so that we can set our base station on top. Okay. screwing it

in. Go ahead and step on the legs. And now I'm going to level this so that the bubble is in the center. And that looks

good. Now, if you look closely here, this little eye piece will allow us to look down and ensure that we are over

the NGS monument. Okay. Now, using a tape measure, I'm going to find the height between the monument and the top

of the trig. All right. And I'm getting looks like 151.8

cm. All right. And so now the last thing to do is attach our base station. All

right. Trying to stay away from the road because of all the noise, but I've loaded up IMLid Flow and I've connected

to our Reach base. So the first thing I want to do is go into base settings, which will allow us to configure our

setup. Right here where it says configure, I'm going to select that. And the coordinate entry method, you may you

want to make sure you select manual. Under antenna height, I'm going to put the height of the base station, which

was 1.518. Okay, save. And then for the position of our NGS point, I'm going to select

choose from project. And then here I input NGS monument. Apply. And there we go. The coordinates for that are here.

So then I can hit save. Okay. I'll go back to receiver and come over to logging. And where it says raw data up

top, I'm going to select start recording. And now it's going to log all of the raw observation files, which will

give us a RIX file that we can export once we're done flying the drone. So now that you've done all of the survey

preparations, you're now ready to pull out the drone and start doing your drone survey. So every drone is a little

different, but the Mavic 3 Enterprise has a button on the back of the battery. You press and you hold

and the drone will power on. And the same thing goes with the controller. Now, you're going to set up

the flight mission along with the proper parameters to ensure the maximum amount of accuracy for your drone survey. Now,

once you load up the controller, you're going to use the DJI Pilot 2 app. So, I'm going to tap on that. So now you're

going to click on the flight route and then you're going to select this little plus sign and select create a route. And

since we're doing a drone survey, we're going to need to create a polygon for where our survey will take place. So I'm

going to select the area route and I can just simply tap where I want this route to be. So I want it to be something like

this. Okay, I like this polygon here. I'm going to go ahead and say okay. And it's going to ask me the drone. So, we

have M3E. That's the Mavic 3 Enterprise. I'll say okay. Now, it's asking me for the flight altitude. I'm probably going

to fly this at, I don't know, 150 ft. I'm going to disable the elevation optimization. I just don't need it. And

then I'm going to go into the advanced settings and ensure my side lap and overlap are both at 80%. I'm going to

change from timed interval shot to distance interval shot. That way, the images that are taken while I'm flying

are more evenly spaced out. And I might actually edit my polygon here so that it covers a little bit more land. So, how

about I go a little bit beyond the borders that I initially set. Okay, I think this might be a bit better. Okay,

we'll go back. And so, now if we look, our mapping area is 4.3 acres. Now, this is a relatively small site. You can

definitely imagine doing a much larger scale project like 200 or 300 or even 700 acres. You can definitely handle

something like this using drone surveying. Everything here looks fine. So, I'm going to go out of this menu.

The next thing I want to do is tap on my camera. And here we go. We have our little camera preview.

And the first thing I want to check is my shutter and make sure it's set to one over 1,000. Next, I want to make sure

our focusing method is AFS. That stands for automatic focus center. And then we're going to be setting our camera

setting to S, which prioritizes the shutter of the camera. If I click on these three dots, I want to make sure

our aspect ratio is 4.3 and we're exporting the images in JPEG. We'll select the three dots here and make sure

that locked gimbal while shooting is on, mechanical shutter is on, and deworping is on. All right. And the last thing I

want to do is come over here and select RTK and ensure our RTK is on. So these are my entry credentials here from

Michigan. And if I go down here, we can see the aircraft has a fixed reading. So our RTK is connected. Now hopefully it

stays connected throughout the whole flight. But even if it disconnects, we set up that base station at the end of

the site on the NGS monument to ensure that we can still run PPK corrections in case we lose RTK in the middle of the

flight. Okay, everything here looks great. I'm going to hit the save button. And now I'm going to take my drone off

of my car and set it on the ground so that we can take off. Okay, so everything looks good here. We have an

estimated flight time of about 5 minutes and 11 seconds. I'm going to hit play, next, and we're going to upload the

flight mission. Ready, and we're going to start. [Applause]

See you later. There it goes. And as we can see here, this is the live view of the drone over the road. So,

it's made it to the end of the site. Here is our tracker on the aerial map. So, it's going to be turning around now

and headed back this way. And the nice thing is you don't have to actually manually control the drone. Everything

is autonomous. The flight mission that you create tells the drone how to fly itself. So, all you have to do as a 107

pilot is monitor the flight and ensure everything is going smoothly. If we take a look, our RTK is strong. We have good

signal, so that's good to see. And there it is. I can see it right above my head. There it is. It's making a loop around

and it's coming back. flying towards us again. There it is. It's right over my head.

You can see You can actually see me in the imagery. And you can actually see how clear the targets are. You can

actually like Oh, wow. I love it. I love how crisp this is. It is 81% complete. So, we're almost done here. It's going

to do one more turn and come back to us and then it will be done. we have nice clear coverage of the road. And so I

purposely created a larger boundary than the park because I wanted to capture extra overlap from all sides of our

project. So this way I know I have all the data that I need when it comes to processing the data and then extracting

it for those juicy juicy deliverables. Okay. And it looks like the drone is done and it's heading back to us. There

it is. All right. And it should begin its descent. [Music]

Welcome home. All right, now that you've finished flying the drone, you have just one more step to do in the field for

your drone survey, and that's to end the session with the base station. I've reconnected to my base station on Imlet

Flow. I'm going to click on logging, and there it is. It's been logging for 31 minutes. That's more than enough. So, we

can go ahead and click stop. So now the log files will be processed and so I can now export them and I can use them for

post-processing kinematic corrections for my drone imagery. So now that you've finished collecting all of the field

data, let's head inside so that I can show you all of the processing and post-processing work you're going to

need to do in order to start selling highquality deliverables. All right, I hope this first part was informative for

you, understanding how GNSS data is collected for our survey control and the process of planning our mission and

flying our drone. Now, before we continue, I want to let you know that on August 8th, 2025, the Survey School is

hosting a free survey research symposium and competition. This live event is going to feature several members of the

Survey School and the research that they've been working on over the last several months. The research

presentations cover a wide range of topics including total station accuracy under various vibration conditions to

analyzing the accuracy of different GNSS corrections to evaluating the accuracy of drone photoggramometry and a variety

of different environmental conditions. We also have a survey school member that is going to be talking about the

education gap and why young college students are struggling to get into the surveying industry. What the trends are

showing us for the future of the survey industry and what we need to do to bridge the gap between the young

generation of surveyors that are entering the industry. This is also a competition. So once the symposium is

over, my social media audience is going to be voting on the best presentation. And the winner of the competition is

actually going to get a free trip to Frankfurt, Germany to attend and present at Intergeo 2025. Interjeo is one of the

largest surveying and geospatial conferences in the world. Interjeo is actually sponsoring the survey school

and so one of our members is going to get a completely free trip to connect with the international community. The

Survey School has lots of scholarships and grants for these conferences. And actually over the last 12 months, six

members from the Survey School got fully funded to attend these conferences absolutely for free. So if that's

something you're interested in and you want to see more of what the Survey School members are up to, then

definitely attend our symposium on August 8th. Check out the surveyschool.com/symposium

to register. Okay, before we move on to the next part of the course, remember that at the end I'm going to show you

how you can get a free IML Reach GNSS receiver just like the one I used previously and how you can get Pix 4D

MATIC, the photoggramometry software that I'm about to use absolutely for free. All right, good luck and I'll see

you at the end of the course. All right. Now that you've flown your drone and you've collected all of this data, it's

time to input that data into your computer so that you can process it, generate outputs, and then create high

value deliverables for your clients. Following the steps in the beginning of this course will make this process so

easy for you. Everything will work seamlessly and you'll have no problem generating high accuracy drone survey

data. Now, the first thing you're going to want to do is take your drone and then for this one, the Mavic 3

Enterprise, there's a little flap here in the back. I'm gonna pull that back and then pull out the SD card where all

of the images are. Here's my SD card. I'm going to put it into a SD card reader. So, it's a full-size SD card.

And then I've got an SD card reader like this one, and I just plug it in. And now I can import the data to my computer. If

you're working on a desktop, you might actually have an SD card slot, so you can easily put your SD card into there

and then import the data that way. The other thing you're going to need are the Rhinax observation files from your base

station. This is going to be critical so that we can process the PPK corrections for our drone images. So, make sure that

you export that so that you can bring it into your computer. And then finally, a coordinate file for all of the ground

control points and checkpoints that we took using our IMLID reach RS3. Now, one of the reasons I always recommend IMLAD

to anyone that's just starting out in drone serving is that you can access Flow 360 on your computer. And this is a

cloudbased project manager that seamlessly transfers the data from your phone over to your computer. It happens

in real time. So, you don't have to actually send anything. You just finish the project, go home, go to Imlid Flow

360, and you can just open up your project. Here we have our drone survey with all of the ground control points

and the checkpoints. So, I can go through everything and see how the survey looks. And from here, I can

export a CSV file, which will allow us to import these control points into our photoggramometry processing software.

I'll come up to export and if you have the free version of Imlid flow, you'll just export all columns. And that's

okay. You're just going to have to open up Excel and filter through and delete unnecessary columns. But if you paid for

the Survey Plus membership, then you can come over to custom CSV and do it right here inside of Imllet Flow. Now, I don't

like to export everything cuz I just don't need everything. But what I do like to export are these columns. I like

to export the name, the northing, the easting, the elevation, the lateral RMS, the elevation RMS, as well as the code.

These options are the most important for you when you import the ground control points and the checkpoints into the

processing software. So, make sure you check all of these options. Okay, I'm going to hit save and export. All right,

and now it saved that into file explorer. And these are the drones images that we captured. We've got a

total of 130 images. The next folder is the reach base folder. So this is where all of the Rhinx data is. If you open

that up, there they are. And then finally, we have the control points. These are the points that you just

exported, the ground control points and the checkpoints. So the first software we're going to work with is IMLID

Studio. Now, you can download IMLID Studio for free. It doesn't cost you anything. There's no subscription and it

just it all comes with your receiver. So you can download IMLID Studio and this software is what's going to allow us to

update our geo tag locations for all of the images using the PPK corrections from that base station. So here we are.

This is Imlid Studio. And so the first thing we're going to select is the drones Rhinax observation file. This

should be inside of the drone image folder. Here's our image folder. Can click here and there is the DJI

observation file. I'll say open. All right. Very good. Next, we're going to pick the bases observation file from the

Rhinx folder. And then inside of the reach base, there it is. That's the reach base observation file. Open. And

what it's going to ask for is our latitude, longitude, and ellipsoid height for the point that we were

occupying. And so that was the NGS monument that we set up over. We're also going to need the height of the base

station, which again we measured and we have. Okay, here we go. I've updated the positions as well as the height of our

base station. Next, we're going to upload our navigation file. This is found inside of the DJI image folder.

So, I'll click on navigation. We go to DJI and there it is, the nav file. I'll select and say open. And the last thing

we're going to upload is the MRK file from our RTK drone. The MRK file aligns the images with the positions of the

drone. So we update the position of the drone and then we see where the image was taken and then we find out at what

time the image was taken and this aligns our temporal resolution ensuring that the right image gets the right updated

coordinates. Again in the drone images we have our MRK file and it says here the timestamp. So we'll say open. Okay,

everything looks great. I'm going to click process and there we go. Oh, looks like we have a lot of fixed readings

here. This looks very good. Very promising. Okay, fantastic. And now you can see here are the trajectories of the

drone in the mission. And we have a success rate of 100% fixed readings using PPK corrections. So, as a survey,

you're going to feel a sense of relief knowing that all of the images have a fixed PPK correction. So, they're all

going to have that high centimeter level accuracy. And when you process the data in the next phase, you're going to feel

much more comfortable with the results that you get because you know that a 100% of the images have a fixed reading.

Okay. Now, the next step is to export the new images. So, it's going to ask us which folder we want to export to. I

like to just select the folder that had the original images in them. So, this is the folder. I'll just select it. And

then it's going to ask us if we want to update the original photos. And you definitely don't want to update the

original photos. It's best to create new images and you should keep the old archive just in case you ever want to go

back to them. If you used an older drone that doesn't have RTK, you can find those original positions that the drone

gave you. Or if you do have RTK, you can actually run both data sets and process them both and compare the results that

you get. So, it's some good statistical analysis data that you can work with. All right. And now we'll select tag

photos. And this is now going to tag all 130 images. So, we're looking good. While these images are being tagged, I'm

going to tell you about the next software we're going to use, and that is Pix 4D Matic. Now, Pix 4D has been

around for over 10 years and is a leader in the photoggramometry processing space. Their processing engine is one of

the best in the world, and the interface of their software is so userfriendly. Whether you're a new 107 pilot or you're

a seasoned surveyor, you should be able to work with this software. It's very easy and intuitive and I'm going to show

you step by step what you need to do in order to generate your high accuracy deliverables. So I've got Pix 4D Matic

open and I'm going to create a new project. We'll call it drone survey and start. All right. So now we are in. Now

the first thing we're going to do is set our coordinate system to match what we used for our ground control points.

Click on this little pencil and I'm going to select my coordinate system. So that's NAD83 Michigan South and NAVD88

using goid 18 and apply. Okay. So now everything has been reproed so that we are on our state plane coordinates and

then it's as simple as dragging and dropping the images. So we are in our image folder here. And as you see these

images on the front of the folder are the original images that came out of the drone. and the ones that are in the

tagged folder. These are the ones that we just processed using PPK corrections. So, I'm going to minimize and drag and

drop that folder and it's going to add all 130 RGB images. So, now that all of the images are in, you can see all of

the 130 images and these blue dots indicate an unccalibrated camera position. Once the project is entirely

calibrated, then we're going to actually get our 3D model. Now, those ground control points that we measured with the

rover are going to help us increase the accuracy of our data set. So, you're going to import them now, and you're

going to geo reference their positions in each of the images. To do that, you're going to select from disk. And

this right here is the CSV file that has all of the points. So, I'm going to say open. And there we go. We have all of

the points that we measured. So, I'm simply going to select apply. And now all of our control points are here. You

notice one through five are ground control points. Six, seven, and the NGS monument are checkpoints. We'll start by

selecting point number one. And in the images here on the right, if we zoom in, we can see here is target number one.

I'm going to actually expand the size here. So, we're simply going to select the center of the target in the first

image. And now, we'll select it here. Select it here. And then one more time here. and then here. You don't have to

do all of the images. You can do just a couple of them. Once you've done a couple of them, I've done six here. I'm

going to head over to point number two. And I'm going to select the center here. Just on a couple of these. It doesn't

have to be all of them. I do recognize that this says number seven. I wasn't paying attention. I just kind of set

whatever targets were in my car. But when you do this, you're going to pay attention. You're going to set the right

target in the right location or or not. It doesn't really matter as long as you have a clear contrast between black and

white or whatever two colors your target has. Point number three. Here we go. Center. Here's the center. All right.

Point number four. Okay. Good. Point number five. Just georreerencing the center position

of the target. Great. And then for the checkpoints, you're going to do exactly the same thing. The only difference is

the checkpoints will not influence the data set. They're not going to improve the accuracy in any way. They're merely

designed to check our accuracy after we've processed the data with the ground control points. So, we'll select select

select. And one more. There we go. And then number seven right there. And right there, here and here. And finally, the

NGS monument. Now, I can see the point underneath the base station. Of course, it's a little difficult to see. Anytime

you're working with a point that's very difficult to identify the center of, I would recommend turning it into a

checkpoint. So, in the case of this survey monument, you know, it's hard to see because there's no contrasting

colors like in a target. So, I'm going to utilize this, but I'm going to use it as a checkpoint just to make sure that

our data comes in relatively accurate. So we'll do like I said before few of these. Okay, I think that looks fine.

Great. So now I can minimize and now we have geo referenced our control points. So now you can calibrate all of the data

and get an initial photogometric 3D model. You're going to select calibrate and then start. And depending on the

size of your project, you may have a larger project. This could take a few minutes to a few hours. So, we'll come

back once the calibration is complete. Lot of boring math later. All right, and the processing is

complete. Let's take a look at our results. So, if we look here, we can see all of the calibrated camera positions,

which are the green dots that are in Pix 4D Matic, and they're shifted slightly from the initial camera position that we

had. And if we want to check our accuracy, we'll come over to reports. And here we can see all of the images

were calibrated. Our ground control points have an RMSSE of 0.047 and the checkpoints have an RMSSE of

0.049. So this is telling you that this data is coming in at 500ths of a foot or just

under 2 cm. And if you're getting results like this, then you should feel really comfortable because you used high

accuracy control on your ground control points and checkpoints. You use RTK corrections to get high accuracy

trajectories on your drone. And on top of all of that, you did PPK corrections on the drone's trajectories to ensure

every image had a high accuracy geo tag so that when you go to process data, it comes in accurate. And this is how you

create high accuracy drone surveys. Now that we have a high accuracy drone survey, we are now going to process out

the point cloud, the mesh file, the DSM, and an ortho image. So you're going to come back to processing options and

you're going to select dense point cloud image prep-processing mesh DSM and ortho and you're going to

start. Now this process will take some time. So again I'm going to leave come back and show you what the results look

like. One eternity later. All right the outputs are done. Let's

take a look and see what we've got. Right off the bat I can see we have our DSM. This is going to show us kind of

the elevation map of our project. Beneath that, we have our ortho image, which is a very highresolution image of

our project. I mean, look at how much we can zoom in. And look how clear that imagery is. That's because if you use a

mechanical shutter camera, you're going to get very crisp and clear imagery. And if you've taken hundreds or even

thousands of images when you calibrate those camera positions together and generate a large 3D model, you'll also

get a highresolution 2D map. And that's what we call an orthomosaic or an ortho image. So very nice, very clear, very

crisp. I love the fact that this came in very nicely. And you can see just how accurate the data is as well. I mean all

these control points. I can zoom right in and see there is the center position of the point. Where is this one? Oh,

there it is. So, this is very crisp and clear. Love it. Love it. Looks great. And then the most important thing here

is what your 3D point cloud looks like. This is what a 3D point cloud is supposed to look like. It's as if you're

actually in the park. It's actually a reconstruction of everything that we did. Very cool. Very nice. And this

takes it a step beyond images and videos. By doing surveying work with your drone, you're able to create

outputs like this, a complete 3D model and reconstruction of an entire site. And obviously, as a survey, this needs

to be highly accurate. So, we've taken those proper steps to ensure the accuracy of our drone survey. Now, this

is all great, right? This is beautiful and it's nice to look at, but you and I both know this is not the final product.

So now I'm going to share with you the secret to give your clients high valuable deliverables. This is what

clients are going to pay you thousands of dollars for. It's not this 3D model. It's not the point cloud. It's not the

ortho image. It's none of that. But it's what you're going to do with these outputs is what's going to create a high

value item. And what I want you to understand is that there is a clear difference between data and information.

Data is what we have here. This is all data, right? There are numbers, there are points, we did an accuracy analysis

on it. Yeah, this is data. Clients are not going to pay you that kind of money for data. They're going to pay you for

information. And in order to give them information, you're going to have to extract it from the data. And this is

what will set you apart from the rest of the competition. I'm going to show you two more softwares that you're going to

need. And one of these products, if you know how to use it and can properly execute the final step, will make you

dominate the drone surveying industry. You'll be able to take this data and provide high valuable deliverables,

giving your clients the information that they need. Now, before we get to that, the first software you're going to need

is an extension of Pix 4Dmatic, and it's called Pix 4D Survey. Now, Pix 4D Survey seamlessly integrates with Pix 4D Matic.

The reason you're going to want to use this software is it's going to simplify the data set that you have here. This

point cloud has millions of points and the file size on this thing is going to be massive. So by simplifying what you

have, you're going to create a manageable data set that you can then give to your client along with

additional information that will be useful for the survey. So it's really simple to take this data from Pix40matic

and put it into Pix40 survey. Up at the top here, you're going to click on file and come down to open in Pix 4D survey.

This is going to automatically take all of the data, all of the images, the outputs, the control, everything, and

bring it into Pix 4D Survey where you're going to extract additional outputs. Okay, we have Pix 4D survey opened. And

here's the point cloud that we were just looking at over at Pix 4Dmatic. And if we look at the point cloud, we've got 16

million points. So yeah, that's a lot of data and no one is going to want to go through all of those points. They're

going to want you to simplify it for them without compromising important information. Sometimes there is noise

that is not supposed to be there. So Pix 4D survey allows us to delete this noise or outlier data to help us clean up the

drawing. So come over to processes and the first option is distant outlier filter. And all you have to do is just

run this algorithm. It won't take long, but it will delete some points. Okay, so this ran and we're good here. Next,

we're going to define the terrain. The terrain is the ground. So, what this filter will do is it will classify the

terrain and declassify any man-made objects or vegetation or anything that's above the ground and put it in its own

non-terrain classification. All right, and the terrain classification is complete. As you can see, anything in

purple is considered non-terrain. So, we have the house right here. We have this fence, the utility poles that were in

the middle of the site, all of these trees, my car that was back here, and even and even check out how good this is

here by the road. You can see it caught my tripod legs. So, and this filter will allow you to now extract information

from the ground terrain. The next step is the grid of points. Now this is what we call a digital elevation model or

DEM. DEM are a set of points with equal distance away from each other or as close to equal distance as possible. And

generally they are used to lessen the amount of points that we have in a data set without compromising the accuracy.

So what you're going to want to check and make sure is that you're only selecting the terrain filter. Now, for

grid type, I like to use the lowass filter, which essentially will look at an area and it'll find the lowest point

for you, and it'll select that lowest point. The reason you want the lowest point is the higher points typically

could be grass or noise or something that's not the actual ground. So, by ensuring that you're selecting the

lowest point, that is actually going to be the most accurate point because that's actually the true ground. So, I

always like to use the low pass. you can use the smart filter. Um, I just find that they give pretty similar results

and the low pass is just a lot easier to work with. So, I want my spacing to be every five ft. And now I'm going to

select make a grid. So, it's going to generate those points for us. And there we go. We've got points

everywhere along the terrain. So, now rather than working with how many points? rather than working with 10

million terrain points, we now have 9,000 DEM points. So that significantly lessens the amount of data that we have

without compromising the information that we're going to deliver. All right. So now the next thing you're going to do

is called feature extraction. Feature extraction allows us to find important information in the data such as

sidewalks, curbs, roadways, any significant features that you would want in a survey. You're going to extract

them now in Pix 4D survey. And now all the way at the bottom here, we have the layers tab. And you can add a new layer.

And so I'm going to call this back of curve. And I have the option to change the color. So, I'm just going to make

this red. All right. It's asking me if this is a terrain layer. And it is a terrain layer because it's on the

ground. So, I will enable terrain layer. Now, we're going to pick the points on the point cloud where we would find the

curve. I'm going to So, I'm going to zoom in and where I would want to start. I want to start here. I'm going to

select the polyline option and I'm going to start picking points along the curb. So these are points that I want to use

and I can see in the images where these points appear. And just like in conventional topographic surveying, you

would actually take your poll and measure each of these points. But what you're doing now is actually doing the

survey virtually on Pix 4D survey. And I want to see I think I've made it I'm going to just pick one more point here.

All right. So there we go. I've drawn our curb line. The next layer I'm going to add is the gutter layer. And we can

change this color to green. And we are going to include this in the terrain. So we'll start by selecting where the

gutter is. And that's at the bottom of the curb. There we go. Looks like this is it. Here's another position. Here's

another one. And you see it's as simple as just picking the points on the point cloud. Okay, there we go. We now have

our gutter line. All right. I'm going to add another layer. I'm going to call this center line. And this will be the

center line of the road. So there we go. I'm selecting the center of the road. And there we go. That is the center line

of the road. This is also a terrain layer. Now I'm going to create another layer and call it concrete. This

concrete layer is going to be blue and I will make it a part of the terrain. And so now I'll just go through my site and

select any areas that I have some concrete. So those approaches that we had were made of concrete. So I'll go

ahead and map them out now. Here's the first one. Okay. And that looks like it ends here. Very nice. And then the other

one is over here. And then it ends over here. Very good. Now, if I don't know what a material is, let's say I don't

remember, I was there, but I don't remember what the sidewalk was made of, I can actually click on the images and

based off of the aerial images, figure out whether it was concrete, ashfelt, whatever it might be. So now I can look

at the images here and this is definitely ashfelt material. So I'm going to create a new layer and label

the sidewalk ashfelt. And I'm going to change this color to I don't know what did we not use yet? Purple. Let's use

purple. And this is a terrain layer. So that starts over here. So I'm going to map the entire sidewalk. There we go.

Okay. And I think this looks good. Actually went a little beyond where I wanted to go, but that's okay. I'm going

to do the other side now. All right. There we go. We're rocking. We're rolling.

We're extracting data. You're getting all the good juicy information. There it is.

Super easy. Just look. Just click. Just give them what they're looking for. All right. That looks nice. That looks good.

And looks like we're going to come here and finish it off. All right. There we go. We've extracted the entire sidewalk.

And you can actually go into the images and kind of take a look and see where that line goes and how it follows the

true sidewalk. So that's pretty nice. Pretty good validation for us while we're doing our feature extraction. Of

course, there are non-terrain points and lines that you want to add in. This includes trees, fences, and other

features like utility poles. So here's how you add all that information in. Again, you'll click add layer. I'm going

to call this one utilities. And this one, we are not going to turn on the terrain layer. However, I'm going to

turn on the entire point cloud so I can see the non-terrain point. And I'm actually going to turn off the terrain

point cloud. Uh, let me add the colors back in. This way, I can easily see our man-made objects in order to assign

points to them. So, for utilities, I want to add just a point. And for the utility poles, so here's the first point

I want to add. We can add one here. And then looks like there's an electric box here. So those are all in. And I can

actually change the names of these. Utility pole one. I'm going to change this marker two to utility pole 2. And

this one I'm going to call it electric trick. So now we've identified what these features are. All right. The other

thing I want to identify are the trees that are on or around the property. So I'm going to add a new layer. I'm going

to call it trees. And we're going to change this over to green. And there you go. Now we can identify those trees. I

may turn on the ground layer just to kind of help me. Yeah. So, this is definitely a tree here in the middle.

I'm going to add a point here. That'll be another tree. Definitely got a big tree right here. So, I'm just going to

take the top of it cuz I don't really care for the elevation. I just want the position of the tree. Looks like we have

another tree right here. I'm going to pick that tree. We got another one right here. So, we'll pick that tree. We've

got one right here. Tree. We've got one right here. We got one right here. So, there's another tree. Here's another one

here. That's a tree. Okay. We've got one tree here. So, we've got a tree right there. We've got another one right here.

All right. So, all this looks fine. I think that's all of them that are on the property. Yep, that looks like it. Now I

am going to use the same layer to show where there are tree lines and those are just the trees that are along the

property lines. I don't have to show every single tree there, just the ones that are on the property I've shown. So

I'll switch over to polyline and I can honestly just turn off the non-terrain points and I can just click away here.

This makes it very easy to identify where the trees are. So I'm just clicking away here on the edge of our

data set. I think I'm going to end it here and then we'll start it up again on this side. Here's a little pocket

where we don't have too much tree coverage. Looks like we're coming around the neighbor's yard right now. And I

think that is actually the end of the tree line. So, we'll end the tree line there. So, there we go. Now we've

identified where the trees are, the line that goes along the property line along with the individual trees that we marked

out. So that looks pretty good. The last thing I want to identify are the fences. So I'm just going to add one last layer

and we'll call it fence. And this is not a terrain layer, so I don't have to worry about that. So here is where the

fence starts. And I'm just going to take this down. Looks like here's another point on that fence. Here's another

point. Okay, now we're starting to see it a little bit more clearly. There's another one. Here's another fence right

there. Looks like this is the corner. And then it just goes into the neighbor's house. So, we'll click there

and end it there. All right. So, there we go. We've got all those features. I could go more crazy. I could map out the

neighbor's house, his garage. Um, but I'm going to try to keep this simple. This is the information that I'm going

to extract from our data set. Okay. Now that you've finished the feature extractions, it's time to generate a

TIN. TIN stands for triangular irregular network. And what a TIN does is it creates a surface for our project. So

all of the points that you've created, all the line work that you just extracted in the terrain layer will be

included in the surface. So I am going to select tin and then I'm going to change this to the smart edge option. We

also want to make sure that you have use terrain layers as brake lines enabled. And then all you have to do is simply

select generate tin. And there we go. We have our tin model generated. We can see a lot of different features here. This

looks really nice. I can clearly tell my curb and gutter line from when we extracted those lines. Here I'm going to

add those layers back. There it is. So, that's very nice that I can see all that. We can also see the sidewalk.

There it is. You can tell where the sidewalk is leading up to the back of the property, where the bridge is. This

is right here is the bridge. So, this is very nice. I really like how the tin looks. It's very clean and concise and

exactly what we need to generate contour lines. The contour lines will help us visualize the terrain and where there

might be lower and higher elevations. So, right here, contours. I'm going to do one foot contours and generate. So

there we go. We've generated the contour lines. If I turn off the tin and turn on the point cloud. So now we can actually

see how this looks. I could tell based off of the road which way the water would run. I can also see here in the

back where there's actually a creek where that creek is. Turning off the non-terrain points, we can visualize

what the ground looks like underneath all of those trees and where the elevation drops most. So, it looks like

here is the drop point. The creek kind of comes over from here. It comes from under the bridge and where these trees

are is where the water runs. You can clearly see now the difference between data and information. This right here is

the information that you're going to provide for your clients. All right. Now, I want to go over the final step,

and that is to export all of your information and bring it into AutoCAD Civil 3D. This is an industry standard

software, so it's important that you export your data into a file format that they will be able to use. You're going

to want to export all of your data as a DXF file. DXF is the file extension used in order to bring in this information to

AutoCAD Civil 3D. So, here we are. I've opened up AutoCAD Civil 3D. And in order to import all of this data, you're going

to type in insert. And we have here the contour lines. We have the layers which we drew in all of the features as well

as the grid of points. So, I'm going to bring them in one by one. Here's the first one. Here's the second one. And

here's the last one. If we take a look, there's all of our data. And you can actually take this a step further by

bringing in the ortho image that you captured with your drone. You can see now I have our highresolution ortho

image along with all of the information we've extracted. And this right here is a high valuable deliverable that you can

charge thousands of dollars for simply by using your 107 along with your survey skills to do drone surveying. I hope you

found this course to be informative and that you've learned a lot about drone mapping and surveying. Before you head

out, I want to let you know about our free access to Pix 4D Matic as well as the IMLD Reach GNSS receivers. Pix40 and

IMLD have been great sponsors of the YouTube channel for many years, and that sponsorship has actually carried over

into the Survey School, providing our students with free access to hardware and software. The Survey School works

off of a leveling system. Members that are heavily involved in our community and provide value, rank up in the school

to achieve higher levels. As students reach higher levels, they begin to unlock resources such as new courses,

new software and hardware, as well as grants to go to conferences for free. And so if you're a highly motivated

member of the survey school, showing that you are committed to learning as much as you can and contributing to the