A Novel Model for Image Classification based on Persistent Homology

Petar Sekuloski, Vesna Dimitrievska Ristovska, Vassil Grozdanov

Abstract: Topological Data Analysis (TDA) is relatively new field of Applied Mathematics that emerged rapidly last years. The main tool of Topological Data Analysis is Persistent Homology. Persistent Homology tracks the topological features of datasets. In this paper we will introduced a novel model for image classification based on Persistent Homology. In the experimental part we the introduced new novel on real medical dataset. Using this model, the classification was improved.

Keywords: Topological Data Analysis, Persistent Homology, Machine Learning, Computational Topology, Image Classification