U-NET with Extreme Learner Approach in Retinal Vessel Image Segmentation

Retinal diseases are a visual disorder that can cause human vision problems, such as blindness and low vision. Early detection of this disease is important to prevent visual impairment. The identification of diseases relies heavily on the accurate separation of blood vessels inside the retina. Additionally, it is beneficial to administer prompt therapeutic interventions for life-threatening illnesses. In ophthalmology, segmentation of blood vessels is a major component of color retinal fundus images analysis for monitoring and diagnosing retinal diseases, like diabetic retinopathy, hypertensive retinopathy, age-related macular degeneration and arteriosclerotic. This process is effective to identify retinal diseases more accurately. This paper helps in segmentation of blood vessels in retinal images as it proposes a new method which is effective in removing noise and makes blood vessels in images clearer using machine learning techniques. This paper proposes the use of U-net with Extreme Learning Approach for blood vessel segmentation of retinal images. Moreover, this approach suites for processing small and unbalanced datasets.