Hi akear,
I think I have already answered this question in the Competition Rules section.
No, I don't think we will do that. You would have to do it yourself if you think that it will be useful.
It should be easily possible to do it using similar methods as we used for detecting the other objects.
There are two reasons why we don't do it:
1. There is no uniform representation for the terrain, using bounding boxes would be not very useful. So we want to give every participant the opportunity to develop what they think is best for representing the terrain.
2. That's a good way for teams to distinguish themselves from other teams and to potentially significantly improve performance
Regards,
Jochen