Published Paper Details:

In this paper, the proposed approach employs a new methodology for segmenting the blood vessels from the retinal images more effectively. There is possibility for improvement in the process of detection blood vessels using segmentation techniques and then for validation classification methods can be implemented. The database of retinal images is taken from DRIVE and CHASE-DB1, these are available in open source for research. In this proposed work, segmentation of retinal images is compared with sobel and canny algorithms and further this is validated using parameters calculated from vesselness methods. The overall process is carried out in two stages. The first stag is to pre-processing of retinal images from database and to convert the color image into grey scale image and second stage will be for classification methods. From the experimental results, it has been observed that the proposed approach provides to decrease process time and increase accuracy in segmenting Retinal blood vessels.