Hi.
My question is not directly related to DJI products, but related to the field work.

Do you guys know of a way to include GCP inside a forest? Without cutting anything, offcourse.

I need to collect mulstispectral and lidar data over forest plots using a Matrice 350 RTK, but I have no idea how locate these plots on the images. Any help is welcome.


Best.  

You have to try to find location where the canopy is open. You could try to climb and put them on top but that's challenging. Over here ( BE ) : I fly when there aren't leaves on the trees and our forests are small.

Yes, putting GCPs where there is a hole in the canopy is an option. There still needs to be a hole and the drone must pass through it more or less vertically.

But in any case GNSS receivers generally have a lot of trouble obtaining solid coordinates under the canopy, so well....

The only viable option for me is to PPK the LiDAR data with a well-established base station, not too far away (<10km), and having a clear view of the sky at least 15deg.
Then you align the multispectral data by shifting the raster if necessary with the reference lidar data (canopy geometry).

There are GCP targets for LiDAR that you can place under the canopy and identify later thanks to the signal intensity on the secondary returns. I have one and quite honestly it's a waste of money and time.

I'm thinking of something, maybe you're looking to know the position of a precise center plot under the canopy. There are two tips that I practice regularly.

The first is to do it with a total station that you place at the edge of the forest in a clear space. This includes that you must have your center plot in direct view.

The second I set markers of different shapes, sphere or pyramid, I have a preference for spheres. You place them on a stake at the location of your center plot, then with the point cloud it's quite easy to find them. It is imperative that they are colored white and that this paint has maximum reflectivity. To try before putting them in the forest. To test you put your sphere next to a piece of wood, you will see if the difference in reflectivity is really marked.
Buy a children's globe and paint it white.

Земной шар PNG (imgpng.ru)

LV_Forestry Posted at 2-8 07:43
I'm thinking of something, maybe you're looking to know the position of a precise center plot under the canopy. There are two tips that I practice regularly.

The first is to do it with a total station that you place at the edge of the forest in a clear space. This includes that you must have your center plot in direct view.

That is exactly what I am trying to do. Given that I work on tropical forests, opens in the canopy or having direct view of the plot from outside the forest is very unlikely.


I will try the globe method!

Thanks very much.  

LV_Forestry Posted at 2-8 07:43
I'm thinking of something, maybe you're looking to know the position of a precise center plot under the canopy. There are two tips that I practice regularly.

The first is to do it with a total station that you place at the edge of the forest in a clear space. This includes that you must have your center plot in direct view.

Have you ever tried one of those aluminium balloon for kids?

AR_AirPrecision Posted at 2-8 13:18
Have you ever tried one of those aluminium balloon for kids?

[view_image]

No i haven't. But there may be one issue by using this balloon, it is the fact that there color is not uniform, and almost impossible to paint in a durable way.

Having uniform color is critical because then the reflectivity value will be +/- always the same.

To found them easily we use the reflectivity filter function. All point related to trees or soil with low reflectivity are disappearing. Iam talking about LiDAR reflectivity here.

Nevertheless with uniform color globe, it remains two issues:
-you can't use it when there is snow because in this situation the reflectivity is high on everything covered by snow.
-It does not have the precision of GCP. It is really made to detect what we name center plot in forest, which are usually used as reference to count how many trees there are in a certain radius. Then with azimuth and distance we are able to detect trees and compare ground measurements (trunk diameter, species...) and drone measurements (spectral Index, tree height...)