Brain tumors detection and segmentation in MRI Images

Abstract

The procedure of detecting brain tumors in a magnetic resonance images (MRI) is a difficult procedure due to the variability and complexity of the data size, data shape and texture of the MRI brain image which consist of lesions. The similarity of intensity between the normal tissue and the lesions leads to special approached in MRI scans. Due to overwhelming cost considerations the need a special approach is required to detect the lesions using single spectral MRI images. In this paper, we have proposed an automatic system which is able to detect the lesion, tumor area. The experimental result demonstrates the success of segmenting brain tumor with high accuracy and less computational complexity, the paper includes the study of comparison of Gabor wavelet with effective feature extarction as a contribution towards tumor segmentation.

Keywords: MR