Home // ICQNM 2018, The Twelfth International Conference on Quantum, Nano/Bio, and Micro Technologies // View article

Brain Tumor Detection and Segmentation Using Hybrid Approach of MRI, DWT and K-Means

Authors:
Parvinder Singh
Mansi Lather

Keywords: Brain Tumor; Image Denoising; K- means Segmentation; Magnetic Resonance Imaging (MRI); Morphological Operators; Skull Removal.

Abstract:
In medical image processing, the detection of a brain tumor is considered as one of the most difficult and time consuming activities. The foremost aim of this paper is to design an efficient algorithm for brain tumor detection and segmentation. This paper presents an improved hybrid method using Discrete Wavelet Transform (DWT), morphological operators, K-means and Otsu's thresholding technique for detecting and segmenting the tumor in the brain. DWT is used for image denoising. Morphological operators are used for removing the skull portions from the brain images and K- means clustering and thresholding approaches are used for image segmentation and finally to detect the brain tumor. The proposed method results in low values of Mean Square Error (MSE) and Bit Error Rate (BER) and high value of Peak Signal to Noise Ratio (PSNR) as compared to existing watershed and region growing segmentation methods and thus outperforms the existing methods.

Pages: 7 to 12

Copyright: Copyright (c) IARIA, 2018

Publication date: September 16, 2018

Published in: conference

ISSN: 2308-3530

ISBN: 978-1-61208-665-1

Location: Venice, Italy

Dates: from September 16, 2018 to September 20, 2018