Estimation of field of view on human fundus image

Quality analysis of fundus image is the main idea of this work. In case of diabetic patients, the surface of a glaucomatous retinal disc can slope steeply into the cup, this may affect the inner anatomy of retinal disc and thus the reduction in the optical disc size. In this project fundus image of any quality is taken and has been acquired into our system, finally deduce if it is a good or a bad quality image. The two parameters used for assessing the quality of the images are non-uniform intensity and stray light. In non-uniform intensity method if the shape match indexing (SMI) value is lesser than 97.5 then the image falls under bad quality. And in the Stray light method, based on the size of the area, if it is greater than 4000 it is concluded that our image has Stray light and it is a bad quality image. In the end comparison and contrast is made based on the parameters. Also, the noises that are present in the image is also brought into focus and eliminated. The results to calculate the accuracy, specificity and sensitivity based on the field of view and resolution is also considered.