Feature - Based Optical Flow Computation

Visual servo can be achieved by extracting motion information from the optical flow, defined as the apparent motion of brightness patterns of an image. Control performance using this technique fully depends on accurate estimation of the flow velocities. It appears that most correct values of the optical flow usually locate at the regions with image features. In this paper, feature-based optical flow computation methods are presented. The approaches of scatter of brightness, edge acquisition, and feature orientation are developed. Simulation studies on six test images were conducted to demonstrate accuracy improvement of the estimated optical flow. Keyword: Image features, Machine vision, Optical flow, Visual flow.