Denoising and Contrast Enhancement of CT Brain Image Using Median Filter and HE

Abstract

In current medical procedures, healthcare professionals possess various supplementary diagnostic instruments that offer immediate, and accurate results. The picture of the anatomical structure of the human body is mostly achieved by using Computed Tomography (CT), and X-ray images. These methodologies are widely employed and considered as essential tools in the field. In recent decades, numerous algorithms have been put forth to assist medical professionals in improving CT images for disease diagnosis purposes. The act of enhancing images has the potential to unintentionally intensify the presence of noise or introduce distortions, ultimately compromising the accuracy of diagnostic outcomes. Consequently, utilizing suitable methodologies to mitigate noise and amplify contrast efficiently is paramount to alleviate the adverse consequences. The algorithm presented in this paper utilizes linear and nonlinear filters, including mean and median filters, to achieve optimal results. The proposed algorithm effectively addresses the Salt and Pepper Noise (SPN) impact on CT brain pictures. This is achieved by utilizing the median filter, which enhances the overall image quality and conserves important structures, such as edges, at various intensity levels. Afterward, the contrast of the CT brain images is improved by utilizing an enhanced median filter. This is achieved by implementing the Histogram Equalization (HE) technique, and it is contrasted with the Adaptive Histogram Equalization (CLAHE) method. The effectiveness of the implemented enhancement techniques can be assessed by comparing their outcomes using the Peak Signal-to-Noise Ratio (PSNR) metric.