"base height field"

[RalfH]

In many cases, people want to model the ground surface below a sparse vegetation cover. Using "arbitrary" geometry construction will result in a model of both ground and vegetation, requiring the user to manually remove vegetation or to apply some kind of vegetation filtering algorithm in an external software. "Height field", on the other hand, results in a model of the vegetation canopy.

It would be very useful to have a "base height field" geometry construction which would result in a model of the ground surface.

[Mohabon]

Bump!

I would love it also! Please check the following attachments.

1) "West field.jpg": red line is profile from PS Pro DEM, green line is profile from LIDAR. As you see, on the right side of the profile there is a small river with slopes and trees and vegetation.

2) "City Limit.jpg": again red is PS PRO Dem and Green is LIDAR. Everything is perfect but the "forest" on the slope of the river channel.

is there any workaround to remove all these greenies and have LIDAR-like results?

[Mohabon]

Bump

[Wishgranter]

No, because Photogrametry is PASSIVE method of obrtaining data, LIDAR is ACTIVE, so it can overcome the problem with trees and few other stuff. Only way is to postprocess the data and delete unvanted areas......

[RalfH]

Wishgranter,

that is not the point in this feature request.

The feature request is about an additional model building option.

The existing options are:

(a) arbitrary: builds a model of an arbitrary object, full 3 D
(b) height field: builds a 2.5 D model representing the TOP envelope of the dense point cloud

The requested option should be:

(c) base height field: build a 2.5 D model representing the BASE (bottom) envelope of the dense point cloud

[JMR]

Have you tried exporting points in *.LAS format and processing them with whatever you used to generate terrains from lidar data?
Some time ago I asked one of my collaborators for a routine that sub-samples a point cloud in this way. It superimposes a regular grid on top of the cloud with the step being in size N times the typical point spacing in the cloud.
Each prismatic cell encloses a number of points usually between 0 and NxN so  if it keeps the one with the lowest z you are likely taking the ground of the sample (you can also choose keep ceiling optionally) ?does it make sense for you?. I could try with this very same dataset for you, so if you liked we could speak about how to get a copy of our program.
Anyway I suppose that much better approaches must be built in Lidar programs such as Terrasolid.
Regards

JMR

[RalfH]

JMR,

yes, I know, that is a possibility. But of course (ant that's why I am asking for this feature) it would be very handy to have this already as an option when creating the model in PhotoScan, so I don't have to convert the file type and process with another software. Also, "height field" (which models the top envelope) is already implemented in PhotoScan, so it shjouldn't be too hard to add "base height field". Just trying to push Agisoft a little to make their software even better...

[JMR]

I guess that you might be wrong. height-field does not model the top envelope, if it was that simple you could just rotate your cloud 180 about x or y and model it as height-field and then rotate again and invert face normals... but my bet is that you would have the very same surface.
Height-field is agisoft's name for one implementation of some of the classic 2,5D triangulation algorithms,  probably Delaunay.

I'm not saying that your proposal is a bad idea, but just wanted to suggest a workaround.

[Alexey Pasumansky]

Hello All,

Probably, new Dense Cloud classification feature with the consecutive mesh generation based on "ground" class will produce the desired output.