Loading...

Design and Development of an Image-based Multivariate Control Chart

Kazemi Kheiri, Setareh | 2017

810 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 50148 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Akhavan Niaki, Taghi
  7. Abstract:
  8. Today we live in an era of continuous technology improvement which results in huge changes in different areas of diverse industries. Among the most recent systems for monitoring and quality control which benefits from high speed, are machine vision systems. The output of these systems, are digital images that can be used for monitoring instead of the original products. Unfortunately due to the computational complexity of data extracted from the digital images, traditional methods lose their efficiency. Therefore, in this thesis, a method is proposed to design a model for the monitoring and control of image-based processes, which uses classification methods, that are capable of classifying the observations into more than two classes, using the concepts of Hotteling T^2’s statistics and control limits. Hereby in this project, despite other developed methods, the observations are classified into more than two classes of “in control” and “out of control”, for there exists a third class which is called the “warning zone”. This class can help to speed up the process of finding the changes which deviate the monitoring parameters from their specified value. In this thesis, four classification methods have been used: Support Vector Machines, Linear Discriminant Analysis, Quadratic Discriminant Analysis and kth Nearest Neighbor which are trained based on T^2 statistics as a criterion to recognize the classes in more than two. The performances of these classifiers are then compared in both the in-control and out-of-control state of a simulated process, which the out of control state is divided into two steps; using and not using the sensitivity rules generated from the ability of model to classify images in more than two categories
  9. Keywords:
  10. Multivariate Control Charts ; Linear Discriminant Analysis ; Support Vector Machine (SVM) ; Classification Algorithms ; Digital Image Processing ; Image-Based Processes

 Digital Object List

 Bookmark

No TOC