Desert Tree Identification and height measurement from LiDAR and Ortho imagery.

[TonyTG]

I am performing a tree census in the desert that needs to identify, measure (height), and photograph Joshua trees including, dead and fallen trees. A "boots on the ground" approach is preferred but can become very costly on large projects. I would like to use remote sensing data to perform this census. I am currently asking our drone contractor for 1' Bare Earth DEM as GeoTIFF and .las files, as well as 3" ortho photos. I can obtain almost any other data if needed.

So, my question is how can I used remote sensing data to perform this census? What additional data will I need? I have tried several methods already and have not identified or measured more than 50% of the trees against a sample area of manual "boots on the ground" survey. I would say what I have already tried but I am afraid that would limit answers. I have a little experience with remote sensing data (mostly topo) but have never needed this level of detail or analysis before.

I know that getting photos is probably a stretch.

[David_McRitchie]

Hey Tony,

It might be worth considering using a Deep Learning workflow for this, particularly if you have data from previous surveys that could be utilised for model training.

The following workflow does a similar thing with Palm Trees so might be what you need.

There are also a lot of Deep Learning Models available on the Living Atlas if you search for .dlpk which may help avoid having to build your own model with training data.

Hope that helps,

David