Enhancement Of Retinal Fundus Images For Ophthalmic Disease Screening

Abstract

Contrast enhancement within the domain of digital image processing is inherently application-specific, and the assessment of image quality is fundamentally subjective. Consequently, the interpretation of an image characterized by superior contrast varies among individuals. A multitude of contrast enhancement methodologies has been conceived, including but not limited to contrast stretching, histogram equalization, and homomorphic filtering. This study delineates the implementation of an adaptive contrast enhancement technique predicated on fuzzy logic principles. Fuzzy logic has been empirically demonstrated to be exceptionally proficient in handling data that embodies ambiguity and vagueness. We have implemented and compare the performance of adaptive fuzzy inference system that ascertains the pixel values of the output based on the contrast metrics derived from the input image. The contrast metric employed in this study is the standard deviation of the image. The empirical results have been compared with pre-existing methodologies for contrast enhancement and evaluated through both visual and quantitative metrics