Two dimensional discrete Wavelet transform and Probabilistic neural network used for brain tumor detection and classification

Abstract

This paper present brain tumor detection and classification using discrete wavelet transform and Probabilistic neural network. The 2D Gabor wavelet (GW) analysis and Probabilistic neural network analysis has been generally used in face identification. Owing to the robustness of GW features against local distortions, variance of illumination and Probabilistic neural network give the proper results of classification. Taking advantage of wavelet transform use in face recognition and its proper outcomes, this Gabor Wavelet approach can be used in other image studies, such as practical images. The proper study of Basic brain images is of huge importance in the early detection of brain inconsistency and disorder, as they provide important detached information in the brain. In this paper, a two dimensional Gabor wavelet analysis application for brain images, for early identification of tumor and a method for brain tumor classification, where images are classified into non-cancerous (benign) brain tumor and cancerous (malignant) brain tumor.