Analyzing retinal fundus images is essential for identifying and tracking various retinal disorders, providing essential information about eye health. The early identification of retinal diseases depends on the accurate extraction of blood vessels. Due to the high accuracy required in fundus images, this paper is going to give a step-by-step method for the mere extraction of blood vessels, particularly when outlining thin and very low-contrast vessels. To improve imagery and lower noise, pre-processing techniques include converting the image to grayscale and then using a Gaussian filter can be applied, respectively. The DRIVE dataset is used to assess the suggested methodology. The suggested method’s average accuracy on the driving dataset was 94.744 percent. The experimental outcomes confirm the proposed methodology’s high accuracy rate and prove the method’s ability to work efficiently with different retinal image samples, indicating a potential for the application’s improved reliability and utility in diagnosing retinal diseases and creating optimal treatment plans.