Genetic approach for obstacle detection using linear stereo vision

Presents a genetic approach for obstacle detection in front of a moving vehicle using linear stereo vision. The key problem is the so-called “correspondence problem” which consists of identifying features in two stereo images that are generated by the same physical feature in the three-dimensional space. The linear stereo matching problem is turned into an optimization task where an objective function, representing the constraints on the solution, is to be minimized. The optimization process is then performed by means of a genetic algorithm. Experimental results are presented to demonstrate the effectiveness of the genetic approach for 3D-reconstruction in real traffic conditions