Multi-level Otsu thresholding and Morphological operations for Fundus Image Segmentation for Diabetic Retinopathy identification

Diabetic Retinopathy (DR) disease is found in eyes of diabetic patients, and become the primary cause of vision loss. Regular screening of retinal images can save vision and avoid the sightlessness situation. Although DR is not reversible disease, but if DR identified in early stages and then treatment can lower down the possibility of vision loss. Ophthalmologists can assess the disease from Optical Coherence Tomography (OCT) or Fundus Images. Assessing DR from fundus images is tedious task or may be inaccurate. Computer assisted diagnosis of DR can provide sustainable assistance to ophthalmologists. Image segmentation has been accomplished by the most authoritative thresholding technique. Retinal blood vessel segmentation is the tedious task using fundus images. This paper includes two main parts, in which the author applied the Multiple preprocessing techniques, then in second part Multi-level Otsu thresholding and Morphological operations for Fundus Image Segmentation has performed and the resultant images are shared in each stage.