Study of Lighting Solutions in Machine Vision Applications for Automated Assembly Operations

• Published in: IOP conference series. Materials Science and Engineering Vol. 26; no. 1; pp. 12019 - 12
• Main Authors: Zorcolo, Alberto, Escobar-Palafox, Gustavo, Gault, Rosemary, Scott, Robin, Ridgway, Keith
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.01.2011
• Subjects: Assembly; Automation; Design of experiments; Feature recognition; Fractional factorial design; Illumination; Inspection; Light sources; Lighting; Machine vision; Materials science; Optimization; Parameter identification; Pareto analysis; Robots; Vision systems
• ISSN: 1757-899X; 1757-8981; 1757-899X
• DOI: 10.1088/1757-899X/26/1/012019

The application of machine vision techniques represents an invaluable aid in many fields of manufacturing, from part inspection to metrology, robot guidance and assembly operations in general. An effective illumination of the working area constitutes a crucial aspect for optimising the performance of such techniques but unfortunately ideal light conditions are rarely available, especially if the vision system has to work within small areas, possibly close to metallic surfaces with high reflectivity. This work aims to investigate which factors mostly affect the accuracy in a typical feature recognition and measurement application. A first screening of a set of six factors was carried out by testing three different light sources, according to a two-level fractional factorial design of experiments (DOE), a Pareto analysis was performed in order to establish which parameters were the most significant. Once the key factors were identified, a second series of the experiments were carried out on a single light source, in order to optimise the key parameters and to provide useful guidelines on how to minimise measurement errors in different scenarios.