An Experimental Investigation of Factors Affecting Repeatability in Diameter Measurements Using Machine Vision

Machine vision technology has recently made significant strides, showcasing its ability to efficiently inspect items under evaluation. This study explores the feasibility of employing machine vision for measuring shaft diameter as a viable alternative to traditional instruments. An experimental investigation was conducted to address this challenge by evaluating how various factors influence the repeatability of shaft diameter measurements captured with a camera under backlighting conditions. An experiment was designed and conducted following the latest international standards to determine measurement repeatability. The study systematically varied parameters such as shaft diameter, shaft-camera distance, camera sensor type, image format, lighting intensity, camera sensitivity and shutter speed. Results highlighted significant factors impacting measurement repeatability and facilitated the development of an empirical model to predict repeatability based on these parameters. The model’s predictions were validated through confirmation experiments. The final outcomes showcased comparable results when measured using a digital micrometer. This research advances the understanding of machine vision capabilities in precision measurements and supports its potential application in industrial settings.