Master Thesis by Maciej Borkowski
Title: Digital Image Processing in Measurement of Ice Thickness on Power Transmission Lines: A Rough Set Approach
University: Department of Electrical & Computer Engineering, University of Manitoba
Date: 12 August 2002
Abstract
This thesis introduces a new system for measuring ice accumulated on power transmission lines. The main objective and difficulty is to gain enough information from two dimensional image of wire covered with ice taken with a simple digital camera and to approximate the actual thickness of the ice covering the wire. The proposed solution combines two methods in a non-standard way. Rough Set Theory is used to process digital images. In the context of filtering digital images an extension of Rough Set Theory called the delta-mesh is applied to wire images. The -mesh is designed to combat noise in selected directions without blocking the important information in a wire image. In the second step, a classical digital image processing algorithm, namely, the Gouraud shading algorithm is used to approximate parameter values describing the dependency between actual and measured ice thickness from the image. The core of the system is based on extraction of information from a bare wire with known diameter to approximate the actual ice thickness. A sample application of a complete system to measure ice accumulated on power transmission lines is given. The contribution of this article is an approach to using rough set theory in digital image processing as an aid in directional oriented image filtering.