So I'm having this project about comparing LiDAR and photogrammetry when it comes to spatial accuracy and resolution. I have generated a point cloud from photogrammetry so now my data consist of 3 point clouds:
1)a terrestrial Lidar point cloud (which will be used as reference)
2) an aerial Lidar point cloud
3) a photogrammetry point cloud
I need to align them, georeference them, clip them to the same extent, extract some objects from them (hard targets like buildings and yard walls and some trees) in order to compare them. I need to compare them by calculating cloud 2 cloud distances, RMSE or MAE error, and also compare the dimensions of the objects (height, width, length). For trees, centroid location of the trees can be used for comparison as aerial Lidar was acquired a year before the 2 other datasets. I would like to do this comparison in the point cloud format and in a DSM after a raster transformation of the objects. Object extraction should be ideally be done in an automatic way using some object detection algorithm.
Any recommendations on how to do that are welcome. As deliverables I need the detailed workflow, the edited data, some tables showing the statistical values of the comparison and ideally a report.
Please tell me what do you think of this. If you can do it and what would you ask for it.