Detection of edge points using entropy.

Title: Detection of edge points using entropy.
Identifier: AAI9020771
identifier: 9020771
Creator: Jeanty, Henry Henrick.
Contributor: Adviser: Joseph Barba
Date: 1990
Language: English
Publisher: City University of New York.
Subject: Engineering, Electronics and Electrical | Artificial Intelligence
Abstract: The problem of detecting edges in digital images is an important one in all areas of image processing. In our particular area of interest, namely digital images of microscopic biological specimens, quantitative information such as area and shape is crucial to the analysis and classification of objects. The location of edges can provide this information while minimizing the amount of data to process. An extensive body of literature is concerned with this problem, and part of this dissertation is a survey of this literature.;Edge detection does not seem to have a general solution. Different types of images demand different approaches to solving the problem. Therefore, we describe the specific types of images we deal with along with the problems and constraints associated with them. We perform an analysis of the entropy operator in the cases of noise-free and noise corrupted step and ramp edges. Finally, the methods used in implementing the operator are described as well as final results.
Type: dissertation
Source: PQT Legacy CUNY.xlsx
degree: Ph.D.