Several application areas like medical, geospatial and forensic science use variety of clustering approaches for better analysis of the subject matter. In this paper a new hierarchical clustering algorithm called Disjoint Tree Based Clustering and Merging, is proposed which clusters the given dataset on the basis of initially generating maximum possible disjoint trees followed by tree merging. Proposed algorithm is not domain specific and be used for both data points and image. For the result analysis, the algorithm is tested on medical images for clustering of the abnormality region (tumor) from brain MR Images. Proposed algorithm was also compared with the standard K-Means algorithm and the implementation results shows that the proposed algorithm gives significant results for tumor extraction. © Springer International Publishing Switzerland 2014.
CITATION STYLE
Vidyarthi, A., & Mittal, N. (2014). Disjoint tree based clustering and merging for Brain tumor extraction. In Smart Innovation, Systems and Technologies (Vol. 27, pp. 445–452). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-07353-8_52
Mendeley helps you to discover research relevant for your work.