A Multi-Objective Enhancement Technique for Poor Contrast Magnetic Resonance Images of Brain Glioblastomas

• Published in: Procedia computer science Vol. 171; pp. 1770 - 1779
• Main Authors: Chandrashekar, Leena, Sreedevi, A
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2020
• Subjects: Brain tumors; CLAHE; Glioblastomas; MRI; Particle Swarm Optimization; Glioblastomas; MRI; Brain tumors; CLAHE; Particle Swarm Optimization
• ISSN: 1877-0509; 1877-0509
• DOI: 10.1016/j.procs.2020.04.190

Magnetic Resonance Imaging (MRI) is primary imaging technique for detection of Brain tumors called Glioblastomas. The quality of the MRI images deteriorate due to the patient condition, imaging procedures, resolution of MRI machines and skills of the technician. Image Enhancement is performed to improve the image quality after acquisition. Most of the enhancement techniques deal with varying the intensities of the pixels; however, the image entropy or edge information is never enhanced. Contrast Limited Adaptive Histogram Equalization (CLAHE) is widely used contrast enhancement technique that modifies only the intensity of the pixels based on the intensity of the neighbouring pixels. In addition, it involves selection of the operational parameters empirically. Hence, there is a need to develop a multi-objective enhancement technique to enhance the image quality in terms of entropy and edge information and not only intensity of pixels. As well as to select the parameters for CLAHE adaptively, without any user intervention. The proposed technique shows improved results for contrast, entropy, peak signal to noise ratio, structural similarity, unique image quality index and mean square error.