Enhanced DWT Filtering Technique for Brain Tumor Detection

Abstract

This paper presents Magnetic Resonance Imaging (MRI) brain tumor detection utilizing Fuzzy C Means strategy with an upgraded noise filtering calculation. A novel technique is proposed to enhance the execution of cerebrum tumor discovery. A new calculation for noise filtering is adapted to extract the correct area of tumor, where execution is enhanced by upgrading the threshold task in wavelet filtering strategy as a preprocessing step. Trial results demonstrate that by utilizing proposed calculation, the filtering procedure gives better execution when contrasted with the current methods. The average value of Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) for Gaussian noise is improved by 40% and 41.06% and for Rician noise, which is 13.73% and 25.39% higher than the state-of-art methods. After filtering, segmentation is done to point out the tumor region. For segmentation, Otsu and FCM methods are adapted here and a comparison is made between these two methods. Experimental results show that Jaccard and Dice coefficient of Fuzzy C Means (FCM) with enhanced filtering is increased by 3.6% and 1.3% compared to the methods available in the literature.

KEYWORDS:

• Image analysis
• Image decomposition and image databases
• Image denoising
• Image filtering
• Image processing
• Image segmentation

Additional information

Notes on contributors

R. Remya

R Remya received BE and ME degrees in ECE and applied electronics in the years 2009 and 2012, respectively. She has been with the Department of ECE as an assistant professor at Arunachala College of Engineering for Women, India. Currently, she is working towards PhD degree. Her field of interest is image processing. Corresponding author. E-mail: [email protected]

Geetha K. Parimala

Geetha K Parimala received the PhD, ME and BE in ECE in the years 2010, 1999 and 1990, respectively. She is working as professor in the Department of ECE in Ponjesly College of Engineering. She has published 15 papers in national and international conferences. Her area of interest includes image processing, VLSI design, medical image processing, optical communication and optical image processing. Currently, she is guiding 10 PhD scholars and 4 candidates have been awarded under her guidance. E-mail: [email protected]

S. Sundaravadivelu

S Sundaravadivelu received his PhD, ME and BE in electronics and communication engineering. He worked as professor at SSN College of Engineering. He has published 29 papers in national and international journals. Fifteen candidates have been awarded under his guidance. His area of interest includes optical communication, optical MEMS, optical signal processing, optical image processing, optical sensors and optical networks. E-mail: [email protected]