EFFICIENT SAMPLING FOR COMPUTER VISION ROUNDNESS INSPECTION

Computer vision systems are the most versatile non-contact inspection systems. However, 100% scanning of discrete industrial workpieces for roundness inspection is time-consuming. As a result, the determination of the sample size for roundness inspection using computer vision ins pection systems is a critical problem. The relationship between a (1 - alpha) degree of confidence and a given estimated range of S-2 (ERS2) based on the chi-square distribution at different degrees of freedom has been used to determine the sample size for estimating the centre o f a circle. Sequential test methods reduce the number of samples requi red. A double specification for sequential tests based on the sequenti al probability ratio test, the risk levels of alpha and beta, estimate d centre and a given tolerance (t) of roundness have been applied to d etermine the quality of roundness. The average number of sample pixels (ANSP) required to determine the quality of roundness has also been d erived.