Open AccessSegmentation of Brain Tumor using Glcm and Discrete Wavelet Transform
Author(s) -
Alpana Jijja,
Dinesh Rai
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.f3812.049620
Subject(s) - artificial intelligence , brain tumor , convolutional neural network , computer science , pattern recognition (psychology) , transformation (genetics) , feature extraction , segmentation , wavelet , feature (linguistics) , grayscale , discrete wavelet transform , computer vision , wavelet transform , image (mathematics) , medicine , pathology , biochemistry , chemistry , linguistics , philosophy , gene
To identify brain tumors at an early stage is a challenging task. The brain tumor is usually diagnosed with Magnetic Resonance Imaging (MRI). When MRI spectacles a tumor in the brain, the most common way of determining the type of brain tumor after a biopsy or surgery is to look at the results of a tissue sample. In this research to detect brain tumors faster and accurately the feature extraction techniques are used to segment the tumor affected area. One of such very effective technique of feature extraction measure is the Grayscale Co-occurrence Matrix (GLCM). This research focuses on the GLCM and Discrete Wavelet Transformation (DWT) technique to detect and label the tumor from an image based on the textures and categorizing it according to a tumor or non-tumor category. The convolutional neural network (CNN) uses these features to improve the accuracy to 91%.