Optimization Of Histogram Equation With The Cukcoo Algorithm to Improve Fundus Image Quatlity

Abstract

This study discusses strategies for identifying Diabetic Retinopathy (DR) using fundus images and the efficiency of image pre-processing techniques to improve their quality. Fundus images in medical image processing often experience problems with non-uniform lighting, low contrast, and noise, thus requiring pre-processing of images to improve their quality. This study evaluates the effectiveness of standard histogram equation techniques and optimized histogram equations with cukkoo optimization in order to choose the best technique to improve fundus image quality to identify DR. The proposed technique to produce better image quality improvements will be tested in several performance metrics, such as NIQE, PSNR, and Entropy. the results of this study, the average PNSR before optimization was 50,8, whereas after optimization it became 49,8239. The average entropy before optimization is 4.5514, while after optimization it becomes 3.8577. The average NIQE before optimization was 3,4046, while after optimization it was 4,73. In general, the results of this study indicate that the quality of the fundus image is better using the histogram equation before optimization than after optimization. In other words, Cukcoo optimization is not suitable for increasing the performance of the histogram equation in improving fundus image quality

Comparison of Image Enhancement Methods for Diabetic Retinopathy Screening

“…Diabetes is a common cause of impaired visual function, and DR is the leading cause of adult blindness. [17]. This occurs when high blood glucose levels injure the blood vessels in the eyes.…”