A Spatiotemporal Method for the Monitoring of Image Data

• Activity Number: 358
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Quality and Productivity
• Author(s): Fadel M. Megahed*+ and Lee J. Wells and Jaime A. Camelio and William H. Woodall
• Companies: Virginia Tech and Virginia Tech and Virginia Tech and Virginia Tech
• Keywords: Change-point Model ; High Density Data ; Image-based Monitoring ; Profile Monitoring ; Statistical Process Control ; Steady State

Abstract

Machine vision systems are increasingly being used in industrial applications due to their ability to provide information on product geometry, surface defects, and/or surface finish. Previous research for monitoring these visual characteristics using image data has focused on either detecting changes within the image or between images. Extending these methods to include both the spatial and temporal aspects of image data will provide more detailed diagnostic information. Therefore, in this paper, we show how image data can be monitored using a spatiotemporal framework that is based on the use of a generalized likelihood ratio control chart. The performance of the proposed chart is evaluated with respect to an image-based profile monitoring technique through simulations and experimental work. The results show that our GLR chart outperforms the profile monitoring method. More importantly, the simulations show that our method provides a good estimate of the change-point and the size/location of the fault, which are important fault diagnostics' metrics that are not typically reported. Finally, we highlight some research opportunities and provide some advice to practitioners.