Identification of Melanoma in Dermoscopy Images Using Image Processing Algorithms

Asha Gnana Priya H, J. Anitha, Jacinth Poonima J.

2018 International Conference on Control, Power, Communication and Computing Technologies (ICCPCCT) · 2018 · 33 citations · 11 references

Concepts

Abstract

Skin cancer is the most common of all human cancers and is always misunderstood with other kind of skin diseases, so accurate early detection of skin cancer is essential. The main objective of this paper is to segment the lesion and identify melanoma from dermoscopy images. A total of 170 dermoscopy images are used in this research. Firstly, the input images are enhanced for better processing then, the lesion portion is segmented from the enhanced image by two methods 1. Otsu thresholding 2. Morphological operations. The descriptive features are extracted from the segmented lesion. The extracted feature values are used to compute the Total Dermatascopy Score (TDS), which is used to find the presence or absence of melanoma in dermoscopy images. Classification accuracy is calculated to assess the performance of the proposed algorithm.

References

11