Abstract

The initiation tasks of Digital image processing (DIP), the image enhancement techniques are used to better represent the image content and make them ready for further analysis. The pre-processed images can be easily handled by the successive steps of DIP such as image sharpening, image restoration, image segmentation, and object recognition. An image is fabricated with the basic picture elements referred to as pixels. Noisy pixels create distortion in the image and that can be suppressed and smoothened using image pre-processing tasks. Bundles of standard preprocessing techniques are there in the field. A framework named Gamma Corrected Gaussian Filtering (GCGF) is proposed in this article for reducing the noise produced by radiographic or Computed Tomography (CT) machines and enhancing the luminance of the captured images by applying histogram equalization and Gaussian filter followed by Gamma correction. The standard image filtering methods such as Mean filter, Weighted-Averaging filter, Minimum filter, Maximum filter, Wiener filter, Median filter, along with Gaussian filter are discussed and compared with the proposed Gamma Corrected Gaussian Filtering (GCGF) framework through the metrics Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) using the chest X-ray dataset of COVID-19 patients.