strong>Miscellaneous Trends Detection of Neovascularization in Fundus Images Using Convolutional Neural Network
Authors:-Associate Professor Dr.F.R.Shiny, Malar, Abishega A G, Anusree A, Christeena Joy A, God Shaly J
Abstract-Visual impairment is one of the major health prob- lems in the world. The main reasons for visual impairment are lifestyle factors and limited eye care resources. Therefore, early screening and timely treatment are the keys to prevent vision damage. This project proposes a detection of neovascularization in fundus images using convolutional neural network. Wiener filter is used for preprocessing. In preprocessing noise is removed from the input dataset. Image segmentation is a critical step in image processing. One of the most common image segmentation methods is fuzzy c-means clustering. Fuzzy c-means clustering methods have a lot of potential when it comes to extracting detailed features from image pixels. Fuzzy c-Means (FCM) clustering is a popular unsupervised learning algorithm. The selected characteristics are fed into the Convolutional Neural Network (CNN) classifier for data classification using machine learning. This CNN classifier model attempts to reduce the number of features in a dataset. Finally, the CNN classification method is used to improve accuracy.
