Lidar to photogrammetry mesh layer comparisons

[Dadaru]

Hello, can someone suggest a methodology behind Lidar cloud results being compared to photogrammetry mesh levels please.

We have arrived at a job ( large ) and the LIDAR results were flown 2 years ago. We have flown our Drone these past days and the photogrammetry mesh is 200-300 mm difference as an average. We are running various software packages and still having a large 'average height difference' between both sets of data.

Is it simply a case that you cannot compare both surveys as it is like apples and oranges? Or is there a software to analyse it further?

I know the question is a bit 'fluffy' but I have a meeting tomorrow and I need to get my head around the 300mm-400mm difference over this 400,000m2 allotment. There is another 5,000,000m2 to further compare or do I just tell the Engineers it cannot be compared?

For example, how do we know the Lidar sensors were of good quality ..... all I have from the Lidar flight is an RMS of 0.64m. A report was not available just a statement of achievement re: the RMS.

Anyways, any suggestions would be appreciated. Thanks.

 

[jaja6009]

What was your RMS?
If your RMS as an example is 2 to 3 cm, the problem isn't your data, it's the older data.

0.64m is 64 cm and is not that high quality.
How many check points were used in the LiDAR data?

You can 100% compare data from different times, dates and data.
I compare my photogrammetry point clouds to authoritative LiDAR data all the time and have excellent results.

If the older data is shifted (And it seems it is) you could shift it in Cloud Compare or other software.

Hopefully our resident expert ADM Geo will chime in.

 

[Dadaru]

What was your RMS?
If your RMS as an example is 2 to 3 cm, the problem isn't your data, it's the older data.

0.64m is 64 cm and is not that high quality.
How many check points were used in the LiDAR data?

You can 100% compare data from different times, dates and data.
I compare my photogrammetry point clouds to authoritative LiDAR data all the time and have excellent results.

If the older data is shifted (And it seems it is) you could shift it in Cloud Compare or other software.

Hopefully our resident expert ADM Geo will chime in.

Thank you for the explanation and suggestions. Knowing that you compare regularly the data of both formats is re-assuring.
Thanks for taking the time to respond. Cheers

 

[Dadaru]

Thank you for the explanation and suggestions. Knowing that you compare regularly the data of both formats is re-assuring.
Thanks for taking the time to respond. Cheers

 

[jaja6009]

I just want to add. You stated comparing mesh to the LiDAR.
Compare the point cloud from your photogrammetry to the LiDAR point cloud.

You can use Cloud Compare, it's free. Shift the LiDAR data to your data.
Show the engineers your RMSE (I hope you used check points). They should be able to understand that your data (If correct) is the more accurate data (By RMSE) and that the LiDAR data is most likely the problem here.

The above is on the assumption that you used an RTK drone and/or Ground Control Points and Check Points. If you did not, this project is looking like it could be troublesome.

 

[Dave Pitman]

For example, how do we know the Lidar sensors were of good quality ..... all I have from the Lidar flight is an RMS of 0.64m. A report was not available just a statement of achievement re: the RMS.

Your project there highlights perfectly why you (and anyone else including the LIDAR survey op) HAVE to prove your work with checkpoints. Understand that RMSE is a mathematical estimate. It can be a very good indicator of precision. But, without an error report on survey-grade checkpoints, is actually pretty meaningless when it comes to PROVING accuracy.

If you do not have the equipment or the knowledge to survey the locations of some points on the ground that are visible in the data. Consider hiring someone to do that. Then, locate those points in your data and determine the difference (error) to those points. Then, see what the error to those points are in the LIDAR surface. And, bingo. You will have your answer as to which is more accurate. And how accurate it is.