Date of Award

11-5-2020

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

First Advisor

Mitra Mirhassani

Second Advisor

Sazzadur Chowdhury

Rights

info:eu-repo/semantics/embargoedAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

Automobiles endure several challenges when operating on the road that can degrade their performance, functionality, appearance, and overall utility. Although, corrosion is very ancient, it is the most dangerous hazard to an automobile. Corrosion can be defined as natural interaction between the metal and its surrounding atmosphere which results in oxidation of metal. This leads to change in metal properties and can be severely dangerous. One of the easiest ways to recognize corrosion is by using visual inspection methods. Visual inspection results are highly dependent on the operator’s way of analyzing corrosion and operator’s experience. Thus, visual inspection method lack standardization and is susceptible to human errors. In this research, an automated digital method is proposed to detect the surface corrosion and estimate the damage caused. The new approach has been designed to work effectively irrespective of the illumination levels, image dis-orientation and variance in rust texture. The proposed method in proven to be 96% accurate. Furthermore, the proposed method is designed in the form of a noncommercial, cloud-oriented app which is efficient, fast, low-cost, low-maintenance and possesses global accessibility.

Available for download on Friday, November 05, 2021

Share

COinS