Edge Detection to Isolate Motion in Adaptive Optics Systems

摘要

The purpose of the edge detection is to differentiate between edges that remain constant in a scene from those that move. The tracking of moving edges is done by using a reference mask generated by comparing the set of edges in each subsequent frame in time to the current reference mask. At initialization, the reference mask is set to be the edges in the first frame. As time goes on, the edges that move are removed from the reference mask, while those that dont are left behind. Only the edges that are consistent in both the current frame and the reference mask are maintained in the updated reference. This prevents the edges of moving objects from being stored in the reference mask. By iteratively updating the reference mask, the algorithm identifies the static sections of the scenes.