Close range three-dimensional position sensing using stereo matching with Hopfield neural networks

In recent years Vision Systems have found their ways into many real-world applications. This includes such fields as surveillance and tracking, computer graphics and various factory settings such as assembly line inspection and object manipulation. The application of Computer Vision techniques to factory automation, Machine Vision, is a growing field. However in most Machine Vision systems an algorithm is needed to infer 3D information regarding the objects in the field of view. Such a task can be accomplished using a Stereo Vision algorithm.; In this thesis a new Machine Vision Algorithm for Close-Range Position Sensing is presented where a Hopfield Neural Network is used for the Stereo Matching stage: stereo Matching is formulated as an energy minimization task which is accomplished using the Hopfield Neural Networks. Various other important aspects of this Vision System are discussed including camera calibration and objects localization.