ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Mammogram Image Segmentation by Watershed Algorithm and Classification through k-NN Classifier

Journal: Bonfring International Journal of Advances in Image Processing (Vol.8, No. 1)

Publication Date:

Authors : ; ;

Page : 01-07

Keywords : Halaricks Texture Features; k-NN; MIAS.;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

This paper presents a novel approach to detect the tumors in the mammogram images based on watershed algorithm. To increase the performance of the classifier, watershed algorithm combined with K-NN classifier is implemented. The gray level co-occurrence matrices (GLCM?S) are obtained from the mammogram images, through the extraction of Halarick?s texture features are classified. American Society of cancer, UK, provides the benchmark data, MIAS (Mammographic Image Analysis Society) database for the validation of proposed algorithm. These images are used for further analysis by classification into three categories using the algorithms. Mammogram abnormalities are found to be detected using the proposed algorithm with the available ground truth given in the data base (mini-MIAS database), the accuracy obtained is as high as 83.33%.

Last modified: 2018-10-26 20:32:45