Abstract

This thesis introduces a multi-feature based texture segmentation and classifi­ cation algorithm. We first present general background on segmentation approaches which can be categorized into three groups : characteristic thresholding and clus­ tering, edge detection, and region merging. According to our texture segmentation objective, we choose characteristic thresholding and clustering along with region merging as our segmentation method which is accomplished by a popular K-mean algorithm applied in the feature space. Fractal geometry is receiving increasing attention as a model for natural phe­ nomena. We present an efficient method for estimating the fractal dimension of image surfaces and show that it performs better than other approaches used in seg­ mentation algorithms in the past. Since the fractal dimension alone is not sufficient to characterize natural texture, we use also texture energy measures and combine them with fractal dimension as a feature pair to describe and segment natural tex­ ture images. Finally, classification follows the segmentation result. Experimental results are also presented. To simulate enviromental effects, we also add noise to the original image and examine the noise endurance of these fea­ tures. It is found that the fractal dimension and texture energy feature pair performs well in characterizing natural texture; segmentation and classification results are also satisfactory even in noisy enviroments.

Notes

This item is only available in print in the UCF Libraries. If this is your thesis or dissertation, you can help us make it available online for use by researchers around the world by STARS for more information.

Graduation Date

1991

Semester

Fall

Advisor

Kasparis, Takis

Degree

Master of Science (M.S.)

College

College of Engineering

Department

Electrical Engineering

Degree Program

Electrical Engineering

Format

PDF

Pages

134 p.

Language

English

Length of Campus-only Access

None

Access Status

Masters Thesis (Open Access)

Identifier

DP0027991

Subjects

Dissertations, Academic -- Engineering; Engineering -- Dissertations, Academic

Accessibility Status

PDF accessibility verified using Adobe Acrobat Pro Accessibility Checker

Download

Share

COinS