Grant: Idea to Innovation

Computer vision (CV) is an important part of automation and robotics, as it allows one to analyze an image and extract useful information for decision-making in complex environments, 3-D manipulations, modern manufacturing of complex parts and autonomous 3D Navigation. *In the recent year CV also became very present in our daily lives in terms of video surveillance and through its fast growing use in the automotive industry. *The objective of this research proposition is to develop a demonstration software to showcase for the industry the outstanding performances of our newly developed grasping algorithm as well as the optimization, improvement and implementation of an image segmentation algorithm still in development. These algorithms are nature inspired, as they are based on the use of the topological properties of Maxwell's equations of electromagnetism by converting pixels into charged particles or dipoles. *The advantage of electromagnetic image analysis is that it can be used in real time and can analyze characteristics that might not be recognized by other techniques. These characteristics include the computation of contour probabilities, long-distance contour interactions, and feature detection independent of the size, resolution and rotation of the image. Finally, the proposed method functions without the need of training data, which should prove to be an advantage in highly changing environments where robots are expected to work. *