We propose a stereo camera self-adjustment methodology based on a correlation algorithm. We focus on stereo cameras in vision systems for vehicles and Unmanned Aerial Vehicles. We therefore focus on changes of vergence angle. Our approach uses the following methodology: (1) feature detection using an edge density image, (2) stereo matching to find the lost camera parameters, and (3) a self-adjustment procedure. We have confirmed that a combination of these procedures recovers the camera parameters by using the textured features typical of outdoor environments, even if parameters are lost because of a change of vergence angle.