Application of genetic algorithms to 3-D shape reconstruction in an active stereo vision system

Abstract

In this paper, a new method for reconstructing 3-D shapes is proposed. It is based on an active stereo vision system composed of a camera and a light system which projects a set of structured laser rays on the scene to be analyzed. The depth information is provided by matching the laser rays and the corresponding spots appearing in the image. The matching task is performed by using Genetic Algorithms (GAs). The process converges towards the optimum solution which proves that GAs can effectively be used for this problem. An efficient 3-D reconstruction method is introduced. The experimental results demonstrate that the proposed approach is stable and provides high accuracy 3-D object reconstruction.