I havent seen any drones that have multi camera rigs that arent using gopro sensors for the cameras. Its not a bad idea, but really for photogrammetry, you will have a much larger improvement from using a larger sensor vs using more small sensors. Of course, a drone flying with multiple cameras will have to make fewer flights since it is getting multiple capture angles in one flight. The downside is that you will get more coverage in one flight at a much lower resolution. For example, here is the same piece of a building captured by a DLSR on the ground about 15 feet from the subject and a reconstruction from a drone that flew about 30- 40 ft away. When you zoom in closely as I did for the screenshot, it doesnt look very nice but if its only going to be used as a backdrop for a design model, it doesnt necessarily need to look nice when you zoom in very much on it. This is why you need to know what the purpose of the mesh is going to be and what level of accuracy is needed. If its just going to be used for visualization, then you can probably get away with a drone with a gopro sensor. If you are going to need higher accuracy, you will either need to fly that drone much closer, or you will need a drone with a better camera or ground photos. Accuracy can be roughly estimated by using the formula R = (Ls x D) / (f x L) where Ls is the longest size of the sensor in millimeters, D is the distance from the camera to the subject in meters, f is the focal length in millimeters, and L is the longest size of the photograph in pixels giving you R resolution in meters / pixel. I would have to look at the distortion of the high rises to know for sure, but if they are glass skyscrapers then the distortion is probably coming from reflections in the glass (same thing happens with water). This can be corrected using a piece of geometry as a reconstruction constraint.
↧