The effects of convolution of a Gaussian function with an image are investigated in this paper on both a qualitative and quantitative level. This paper studies a methodology of segmentation utilising Gaussian blurring in addition to evaluating the generally known Gaussian-blur in image filtering. Noise is an unavoidable part of the acquisition process. As a result, knowing the impacts of a filtering technique is critical for selecting the suitable technique to effectively filter the image, as the segmentation process can be costly and time-consuming. It’s usually preferable to have an automatic segmentation method that saves time and human labour. In order to analyse the impacts of the convolution in a quantifiable method, we chose a Quality Index to measure the filtering properties. The Gaussian Blur approach should be used in photographs with a lot of noise and a small variance Gaussian function, whereas a higher variance Gaussian function should be used in images with a lot of noise and a large variance Gaussian function.