Sgeyeuk, Kirkham, Blackpool (2026)

08/09/2026

What is RMSE in Photogrammetry?
If you work with drone photogrammetry, you’ll often see the term RMSE – Root Mean Square Error.
But what does it mean or tell you about the quality of your survey?
The RMSE is a statistical measure of how closely your model agrees with known or surveyed positions.
For example, a Ground Control Point (GCP) is tagged in the imagery and assigned coordinates. Once processing is complete, the software compares the modelled position against the known position and calculates the error.
Check Points (CPs) work in a similar way, but they provide an independent assessment of the finished model, because they haven’t been used to constrain the processing.
How is RMSE calculated?
Imagine a CP has been surveyed by GNSS: Known: E 500.000 m | N 200.000 m | Z 100.000 m Model: E 500.012 m | N 199.994 m | Z 100.008 m
The residuals are, Easting: +0.012 m, Northing: −0.006 m, Height: +0.008 m
RMSE squares the errors, calculates their mean, then takes the square root: RMSE = √[(e₁² + e₂² + … + eₙ²) / n]
Why square the errors? Because positive and negative errors shouldn’t simply cancel each other out.
What does RMSE tell you?
If your independent CP assessment produces: Horizontal RMSE = 0.018 m, Vertical RMSE = 0.025 m.
This indicates how closely the model agrees, statistically, with the independently surveyed points.
But an RMSE of 25 mm does NOT mean every point is accurate to ±25 mm. Some points may have less error, while others may have considerably more.

That’s why RMSE should be considered alongside individual residuals, maximum errors, error distribution, outliers, systematic bias and the quality of the survey control.
The important takeaway
A good-looking orthomosaic doesn’t automatically mean you have an accurate survey. Rather than asking: “How accurate is the drone?” A better question is: “How was the accuracy of the final dataset assessed and demonstrated?”
Professional photogrammetry isn’t just about collecting data. It’s about being able to demonstrate the quality of the results.

04/09/2026

When you map areas of dense vegetation any gap will do.

03/09/2026

Accuracy vs Precision – What Are You Actually Paying For?

One of the questions I’m often asked is: “How accurate is a drone survey?”
and while it sounds like a simple question, the answer isn’t as simple as giving a number.

You see there’s an important difference between accuracy and precision.
Accuracy is how close your measured position is to the true or known position.
Precision is how closely repeated measurements agree with each other.

You can have a dataset that is very precise but still not accurately positioned. This is where things like Real Time Kinematics (RTK), Post Processed Kinematics (PPK), Ground Control Points (GCPs) and Check Points (CP’s) become important.

Just because a drone has RTK does not automatically mean that every point in the final survey is accurate to the quoted GNSS specification. The accuracy of the final deliverable depends on the whole workflow from the Drone, Sensor, GNSS, Ground Control, Flight Planning, Image Capture, Processing, QA to the Final Deliverable.

Check Points are particularly useful because they provide an independent way of assessing the accuracy of the finished dataset.

This is where measurements such as RMSE (Root Mean Square Error) can help quantify the positional error. But in photogrammetry this a bit the software marking it own homework but ill cover this in another post.

So, when you're commissioning a drone survey, perhaps the better question isn't simply:
“How accurate is your drone?” Instead, ask, “How do you verify the accuracy of the final survey?”

Because ultimately, you're not paying for a drone to fly over your site. You're paying for data you can rely on.

01/09/2026

LiDAR Doesn’t see though trees, it sees between them.

I’m as guilty as anyone, when I’m asked about LiDAR or to explain how it works I will aften say that it sees through the canopy but it not true.
It’s one of the biggest misconceptions about LiDAR and it make this tech seem almost mythical.
LiDAR does not see through vegetation. But the pulse of light emitted from the unit travels between obstacles giving it a significant advantage over traditional aerial photography when working in areas with vegetation.
Here’s why:
Photogrammetry relies on visible light and images.
If the ground is hidden beneath a dense canopy, the camera generally cannot capture it. The result can be a model dominated by the vegetation, with little or no reliable information about the ground beneath.
LiDAR works differently.
The emitter sends out thousands of laser pulses. Some pulses hit the upper canopy, some hit branches or vegetation lower down. Some can pass through gaps between leaves and branches and reach the ground. They are then reflected, back to the receiver.
The important point is that LiDAR pe*****tes vegetation better, but it doesn't see through it.
This means that in woodland, scrubland and heavily vegetated areas, LiDAR can often produce significantly more useful ground information than photography alone.
But there are limits. In areas with dense, completely closed canopy, thick undergrowth or multiple layers of vegetation the light pulses can still be prevented from reaching the ground or returning to the receiver.
So, when someone says asks if LiDAR can see through trees, you should simply say that “LiDAR has a greater ability to pe*****te vegetation and capture ground returns through gaps in the canopy but does not see through it.”
This distinction is important when deciding which technology is appropriate for a survey because LiDAR isn't magic, it's another measurement tool. The quality of the final survey still depends on flight planning, point density, sensor performance, GNSS positioning, ground control/check points, processing and, most importantly, the environment being surveyed.

27/08/2026

Today was ment to be a LIDAR day but the good old English weather had other ideas.

25/08/2026

Hi everyone.
I'm still new to the video making and this is the second in a series of video's I wanted to make showing the processes I use when drone mapping, from planning to processing.
Now I realise I have a face for radio but id appreciate any help you can offer so take a look and let me know what you think.

https://youtu.be/TWw5AJgN6tI?si=WTKuRGIqeRZSpEaD

20/08/2026

How Accurate Should Your Check Points Be in Drone Mapping?
One of the most important questions in a drone mapping project that is often overlooked is how accurate are your Check Points (CPs)?
You see GCP’s and CPs are physically the same thing, a mark on the floor that is independently surveyed. The difference is the GCP’s are used to process the date and CPs are used to validate it or test how accurately the finished model represents the real world.

But what is an acceptable accuracy?
There isn't one universal number. The required accuracy depends on the purpose of the survey, the specification you're working to, the equipment used and the accuracy required by the client.
A good rule is that your Check Points should normally be significantly more accurate than the accuracy you're trying to demonstrate.

How is CP accuracy measured?
After processing, the coordinates of each Check Point are compared with its independently surveyed coordinates and the differences are calculated in, Eastings (X), Northings (Y) and Elevation (Z)
These differences are then used to calculate the RMSE (Root Mean Square Error) for the dataset.

For example, if your survey is intended to achieve approximately ±50 mm accuracy, you wouldn't want your CP survey itself to have an uncertainty of ±50 mm. Your control and check-point observations need to be considerably better than the accuracy you're trying to verify.
Why does this matter?
A beautifully coloured orthomosaic doesn't automatically mean the survey is accurate. You need independent evidence that the model is correctly positioned and that its measured coordinates agree with known ground positions. That's where well-surveyed Check Points become extremely valuable.

The key question isn't simply "What accuracy does my drone achieve?" It's: "Can I demonstrate the accuracy of my final deliverable?"

At SGEyeUK Ltd, we believe accuracy should be measured, checked and reported — not assumed.

18/08/2026

its been a long day but finally got to processing done and what a job 200 acres and 2.5billion points.

18/08/2026

What Should a Client Ask for from a Drone Survey?
I’ve a few posts now about what drones can offer clients but to be honest I don’t think I’ve covered what a client should ask from a provider even though I’ve been ask many questions from clients before so, for all those considering using drone before booking a drone survey remember it’s only the first step.

One of the biggest mistakes I see is focusing on “Can you fly the site?” rather than asking “What data do I actually need?” so before commissioning a drone survey, there are some important questions to ask:
What accuracy do I need?
Is the survey for general visual information, earthworks, setting out, measurements or engineering purposes? The required accuracy can significantly affect how the survey is planned and captured.

What coordinate system should the data use?
Make sure the survey is delivered in the correct coordinate reference system and vertical datum for your project otherwise the data is of no use.

What deliverable do I need?
A drone survey can produce much more than aerial photographs so depending on the project, you might require:
Orthomosaic, Point cloud, DSM, DTM, Contours, CAD data, 3D model, Stockpile volumes, Site progress information

How will the data be used?
Will it be imported into CAD? Used by a surveyor? Compared against previous surveys? Used to calculate volumes? Shared with a client or design team?
This is important because the intended use should influence how the survey is planned and what data is collected.

A professional drone survey isn't simply about putting a drone in the air and taking photographs. The flight, camera/sensor, GNSS, ground control, flight parameters, processing and quality assurance all need to be considered against the final requirement.

The better the client requirement is defined, the better the survey can be designed to deliver the right information.
So before your next drone survey, don't just ask: “How much does it cost?” Ask: “What exactly will I receive, what accuracy will it achieve, and is it suitable for what I need to do with the data?”
That conversation can make a huge difference to the value of your survey.

13/08/2026

Drone surveying isn't just about the drone.
It’s easy to look at a drone survey and think the aircraft is the most important part. In reality, the drone is just one part of a much bigger system. For professional surveying and mapping, the process can look something like this:

Drone; The platform needs to be appropriate for the site, conditions and mission.
Sensor; The right payload is critical. RGB, survey-grade cameras, LiDAR or thermal sensors all provide different types of information.
GNSS; RTK, PPK or other positioning systems help establish accurate camera positions and improve the quality of the resulting data.
Ground Control; GCPs and check points can provide additional control and, importantly, an independent way of checking the accuracy of the final results.
Flight Planning; Altitude, overlap, ground sampling distance, terrain, obstacles, lighting, weather and the required deliverable all need to be considered before the drone leaves the ground.
Processing; Hundreds or thousands of images don't automatically become a useful survey. The data needs to be processed correctly to create products such as orthomosaics, point clouds, DSMs, DTMs or 3D models.
Quality Assurance; This is where problems can be identified before the data reaches the client. Coverage, image quality, control points, accuracy and the resulting dataset all need to be checked.

Final Deliverable; Ultimately, the client isn't buying a drone flight. They're buying useful, reliable information that can help them make decisions. That's why professional drone surveying is about much more than owning an aircraft and knowing how to fly it.
It's about understanding the complete workflow — from planning the mission to delivering and standing behind the results.

The drone is the tool.
The process is the service.
The data is the deliverable.

Sgeyeuk

08/09/2026

04/09/2026

03/09/2026

01/09/2026

27/08/2026

25/08/2026

20/08/2026

18/08/2026

18/08/2026

13/08/2026

Address

Telephone

Website

Alerts

Contact The Business

Shortcuts

Share

Category

Listing takedown request