An integrated vision system based on coining algorithms

This article describes an integrated computer vision system that is based on the principle that improved function can better be achieved by combining weak and idiosyncratic operators than by striving to “fine-tune” a single operator. This binding or fusional principle emerged during the past decade of research on organic systems. The system combines many of the attributes of a visual image to segregate three-dimensions in a way that overcomes the deficiencies that have long been known to hamper computer image processing. The system can be applied to object detection and recognition, autonomous control, and manufacturing inspection.