Automated roundness inspection

Includes bibliographical references (pages [76]-79)On-line automated visual inspection for quality and process control is becoming a very important requirement in an automated manufacturing environment. In order to meet this requirement, efficient algorithms and methods have to be developed for each specific problem since there is no unique way to solve all problems. The purpose of this thesis is to develop an efficient algorithm to analyze the roundness of a production pan using a general-purpose image processing system. General procedures involved in digital image processing and analysis are first reviewed and discussed. Some related algorithms and methods are modified and applied to the problem, such as edge detection, thresholding, etc. The binary search method is applied to the edge detection to shorten the processing time. Different methods in measuring out-of-roundness are discussed. A roundness measurement algorithm using sampling approach is proposed. Proper sampling plan and sampling size are also discussed and tested. All algorithms and methods proposed in this thesis are tested using real image. The experimental results show that the algorithms and methods developed in this work are both practical and successful.M.S. (Master of Science