CloudCompare forum

Dear Daniel,
sorry for such new post that for sure include some trivial questions however I can't find a good solution by reading other posts.
I get in trouble when I use point clouds to generete raster grid with the rasterize tool.
As an example, I use a very dense UAV-based point cloud (saved as .las) of a small muddy island in a channel made with photogrammetry processing.

https://www.danielgm.net/cc/forum/downl ... ebbecc6107

I'm trying to conver it into a raster grid with rasterize tool but if I use a small cell grid (e.g. 0.05) the generated raster seems OK in CC but when I import it to QGIS I only see isolted pixels with a very low density. Also the hillshade genereted into CC is a very flat gray image without shadows (see the attached screenshot).
https://www.danielgm.net/cc/forum/downl ... ebbecc6107
https://www.danielgm.net/cc/forum/downl ... ebbecc6107
https://www.danielgm.net/cc/forum/downl ... ebbecc6107

I noticed that by increasing the grid size to 1, the result seems much better as you can see, but in this case I lost too much resolution for my puroses.

https://www.danielgm.net/cc/forum/downl ... ebbecc6107
https://www.danielgm.net/cc/forum/downl ... ebbecc6107

So I wondering if there is a good solution to convert an high resolution point cloud into a grid with the same high spatial resolution.

Morevoer, when I import the .las file into Qgis the spatial reference it's ok but the raster created in CC are missing of spatial reference. I tried to change different options of global shift/scale but I get always a raster without spatail references.

Thanks in advance for your support.

Best regards,
Daniele

I see that while you have a hell lot of cells with a low resolution (42 million), you only have about 130 000 non empty cells... Which means most of the cells are empty.

This explains why you see the raster as being very sparse in QGis, and the hillshade is flat as well.

Basically, you don't have enough points to generate such a fine raster. You could 'interpolate' the points to fill the empty cells, but it won't be much better than a coarser grid...

Thank you so much Daniel for your fast and great support.
Do you think that maybe another acqusition using a lower flight altitude and more images can increase the density of my point cloud enough to get a good raster at 2 cm scale? Based on your espertise which is a good density of points?

Best regards,
Daniele

Ah, I hope the person conducting the flight has this experience as well ;)

But yes, that's generally how you obtain a denser point cloud. Or you need a scanner with more beams and a greater frequency I guess.