Automatic detection of atypical pigment network using texture segmentation

Author

Sruthi Chandana Bhavanam

Abstract

"Software was developed to calculate texture features in skin lesion images with the goal of automatically finding atypical pigment network in these images. The software is capable of calculating eleven texture features over an area (window) of specified size. The software can slide a window of specified size over the image, creating a new image where the pixel value represents the texture over a window centered at that pixel. Output images can thus be formed for all the specified features. The maximum and minimum values of texture features of a calibration set of images were used to scale the resulting texture images. A texture segmentation method where the entries in the gray level co-occurrence matrix were summed and then divided by the number of non-zero entries in the co-occurrence matrix gave the best segmentation for atypical pigment network within a set of 28 malignant melanoma images. This feature found at a pixel distance of 22, for a window size of 41 gave the best segmentation in the set of melanoma images. When applied to a set of benign lesions, false positive atypical pigment network areas are segmented in approximately 50 percent of the images"--Abstract, page iii.

Advisor(s)

Moss, Randy Hays, 1953-

Committee Member(s)

Stanley, R. Joe
Shrestha, Bijaya

Department(s)

Electrical and Computer Engineering

Degree Name

M.S. in Computer Engineering

Publisher

Missouri University of Science and Technology

Publication Date

Fall 2008

Pagination

ix, 97 pages

Rights

© 2008 Sruthi chandana Bhavanam, All rights reserved.

Document Type

Thesis - Citation

File Type

text

Language

English

Subject Headings

Image processing -- Computer programsMelanoma -- Diagnosis -- Computer programsSkin -- Cancer -- Diagnosis

Thesis Number

T 9426

Print OCLC #

313462890

Recommended Citation

Bhavanam, Sruthi Chandana, "Automatic detection of atypical pigment network using texture segmentation" (2008). Masters Theses. 69.
https://scholarsmine.mst.edu/masters_theses/69