Hello Everyone,
I'm trying to create a high resolution (5cm GSD) ortomosaic out of a few hundred images flown at 120m AGL with a DJI Mavic 2 Pro.
I get a reasonably looking multi million point dense cloud that I use to compute a DSM of my area of interest.
Then I compute an orthomosaic based on the DSM (trueortho) which leads to a beautifully looking result everywere excepte on the edges of the buildings where I get jaggy edges that render the orthomosaic pretty much useless as it is.
I know I can recompute the orthomosaic using a DTM instead of a DSM but then I would have to invest time lying polígons to manually choose which pictures are best looking, something I would like to avoid.
Is there a way to filter out the façade points out of the high density point cloud (it seems that the noise on the edge of the buildings mostly comes from poorly matched points on the facades)?
If my question does not go in to the right direction to solve my jaggy building edges problem and you have a suggestion I have bested interest in hearing your thoughts!
Isaac