Only recently I got into using UAV for mapping. We focus on orto/DSM generation for grasslands and agricultural crops, hence monotonous (green) landscapes.
During the last project, the mosaicing was not satisfactory, and I guess the main cause was due to differing lightning conditions. The sky was sunny during the entire flight, so it was not due to the shadowing caused by clouds. Some people in the sector of manned aerial photography warn about white 'hot spots' appearing on photos in case the sun is too high, they suggest sun angle between 30 and 60 degree.
For my case, however, I noticed that the photos taken during flight lines with the sun ahead had different lightning conditions than the photo during flight lines with the sun in the back. As the landscape is monotonous green (but also with stones, ditches, etc.), it logically complicates the mosaicing processes. During earlier flights, when sky was clouded, this type of landscape gave very good ortofotos.
Hence, I am wondering if someone else noticed this kind of lightning influences? Do you take the position of the sun into account with respect of the orientation of the flight lines? What is your general mission planning strategy for obtaining good results (timing? cloudiness? angle sun?).
I also wonder whether there are persons using perpendicular flight lines (double overflow) for mapping difficult terrains (monotone landscapes, urban area, etc.).
Many friendly greetings
I find only using images from one direction helps. I try and fly at midday to reduce shadow and with as much light as possible.
Hey Gary, thanks for the quick reply. In case you fly in one direction, don't you lose much of the battery energy when after each line going back to the beginning for making the second line? I was thinking that maybe flying 'perpendicular' on the sun direction would help to diminish effects. Greetings.
Hi Wally, it might have something to do with the bi-directional reflectance (BRDF) as you acquire imagery during both forward- and back-ward scattering orientations. It has everything to do with the relative position between the sun and the observer (i.e. the camera sensor). To learn more about this, check out: http://www-modis.bu.edu/brdf/brdfexpl.html
There is also some research conducted on how to best correct for BRDF with UAV imagery. See: http://www.mdpi.com/2072-4292/2/3/819