Culminating Project Title
Date of Award
Culminating Project Type
Computer Science: M.S.
Computer Science and Information Technology
School of Science and Engineering
Jie Hu Meichsner
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Keywords and Subject Headings
Object Orientation, Classification, Regression, Principal Component Analysis
Product validation is a manufacturing step in which a product undergoes safety and functionality tests before it is released. Presently, human labor is used in product validation. Automating product validation can alleviate errors caused during the validation process by human error. However, an impediment to automated product validation is the orientation of the product. It is crucial that a product is correctly oriented before the artificial system can run the quality check. This experiment is designed to address the object orientation problem encountered during the object recognition step of the product validation process. Three techniques were examined to solve the object orientation problem. The first is Principal Component Analysis, and the other two are supervised machine learning techniques: Classification and Regression. The results of the three methods are surveyed, and the limitations of the methods are discussed. The machine learning models are better suited for solving the object-orientation problem in comparison to Principal Component Analysis. The classification model made better predictions twice as often on average than the regression model.
De Silva, Amila, "Object Orientation Detection and Correction using Computer Vision" (2020). Culminating Projects in Computer Science and Information Technology. 33.