Vision-based shape recognition and analysis of machined parts

Machine vision has the potential to significantly impact both quality and productivity in computer integrated manufacturing, due to its versatility, flexibility and relative speed. Unfortunately, algorithmic development has not kept pace with advances in vision hardware technology, particularly in the areas of inspection and decision making. This paper deals with the development of machine vision algorithms for automated inspection of production parts. The inspection system presented in this work consists of three parts in series: segmentation, recognition and analysis. The input of this system is a set of ordered boundary data extracted from the object, and the output includes the identity of this object, and its pose, dimension and out-of-profile error, Computer experiments have shown the proposed algorithms to be consistently accurate and extremely fast. These algorithms can be easily programmable lo inspect different types of shapes, which makes the vision system generic and flexible. Furthermore, these algorithms were developed based on the current definitions of dimensioning and tolerancing standards provided by ANSI YI4-5M-I982, so that the results generated by the system are unique and interpretable.