Noise removal and contrast enhancement in fundus images via morphological operations

Retinal images or fundus images are used to support the diagnosis of eye diseases and cardiovascular diseases. However, if these images present poor quality such as noise, low contrast or poor details, they may not be useful for diagnosis, especially in automated image analysis systems. In this paper, a scheme for noise reduction and contrast enhancement of fundus images is presented. This approach is based on mathematical morphology operations and consists of two steps. In the first step, the Open-Close Close-Open (OCCO) morphological filter is used to smooth the image. In the second step, a contrast enhancement algorithm based on the top-hat transform is used. The proposed approach obtains better results in terms of noise reduction and contrast enhancement compared to state-of-the-art methods.