3D measurement strategy based on a projection invariance motion analysis with an artificial retina sensor

A 3D measurement strategy of an object using a specially designed artificial retina sensor (ARS) is presented and a simple template matching method on the polar domain with a camera motion constraint is described. The proposed artificial retina sensor resembles the human retina such that the receptor density increases towards the focal point at the center of the visual field. In addition, ARS accesses the input image directly in the polar domain so that ARS functions like the striate cortex at the visual field in the human brain. The variance of object contours occurs only horizontally, in the polar domain as the ARS moves towards or away the object. Thus a simple horizontal edge detection operator is applied to the moving object data and the 3D information is obtained from the optical flow of the consecutive image frames. Here a template matching technique is used to the horizontal axis with image data attained from the ARS in real time so that this technique may be applied to the high speed applications such as the real time manufacturing procesess and the navigation robots