A stochastic optimization approach for roundness measurements

In this paper, we develop a vision-based inspection system for roundness me asurements. A stochastic optimization approach has been proposed to compute the reference circles of MIC (maximum inscribing circle), MCC (minimum cir cumscribing circle) and MZC (minimum zone circle) methods. The proposed alg orithm is a hybrid optimization method based on simulated annealing and Hoo ke-Jeeves pattern search. From the experimental results, it is noted that t he algorithm can solve the roundness assessment problems effectively and ef ficiently. The developed vision-based inspection system can be an on-line t ool for the measurement of circular components. (C) 1999 Elsevier Science B .V. All rights reserved.