Masters Theses


"Malignant melanoma is one of the deadliest forms of skin cancer and is one of the most rapidly increasing cancers in the world. Image analysis techniques for the early detection of melanoma are dependent upon the detection of multiple color shades in melanoma images. Use of magnified visible-light imaging by dermoscopy allows detailed investigation of these color shades. This research explores the fuzzy logic methods for the detection and characterization of blue color shades to discriminate melanoma from benign skin lesions. Standard thresholds are placed on the RGB dermoscopy images firstly and then the restrictiveness is explored by implementing fuzzy logic. Multiple shades of blue (lavender blue, light blue and dark blue) are segmented based on applying alpha cuts, for different combinations of color fuzzy sets, whereby, a pixel is included in the segmented blue area mask if the pixel satisfies the alpha-cut membership constraints for all of the color fuzzy sets. An accuracy of 82.7% was obtained at alpha cuts of 0.30, 0.40 and 0.65 from the features derived on an 866-image dataset using logistic regression model of Statistical Analysis Software and an accuracy of 81.4% was obtained at an alpha cut of 0.25 using the SVM classifier"--Abstract, page iii.


Stanley, R. Joe
Stoecker, William V.

Committee Member(s)

Moss, Randy Hays, 1953-


Electrical and Computer Engineering

Degree Name

M.S. in Computer Engineering


Missouri University of Science and Technology

Publication Date

Spring 2013


vii, 30 pages

Note about bibliography

Includes bibliographical references (pages 27-28).


© 2013 Mounika Lingala, All rights reserved.

Document Type

Thesis - Open Access

File Type




Subject Headings

Melanoma -- Diagnosis
Pattern recognition systems
Image analysis -- Mathematical models

Thesis Number

T 10633

Electronic OCLC #