Publication Type : Book Chapter
Publisher : Taylor & Francis Group
Source : Machine Learning and Deep Learning Techniques for Medical Science
Campus : Nagercoil
School : School of Computing
Year : 2021
Abstract : Glaucoma is the main disease that injures the optic nerve of the eye and causes vision loss. In the medical field, retinal image analysis is broadly used to identify the irregularity of the eye. Thus, glaucoma detection is mandatory for the previous stage. Accordingly, the segmentation process is employed to correctly automate glaucoma analysis. In order to correctly detect glaucoma disease for minimizing the time, Nonlinear Teager-Kaiser filterativeinfomax boost clustering-based Image segmentation (NTKFIBC-IS) technique is proposed. NTKFIBC-IS technique comprises three major processes, namely preprocessing, segmentation, and feature extraction. Then, the input fundus images are used to perform image preprocessing. Then the noise is removed and obtained quality improved image by applying Nonlinear Teager-Kaiser filtering technique. The input images are split into diverse segments by applying the infomax boost clustering technique. Finally, the region of interest is chosen and different clinical features are extracted. The statistical analysis is carried out for identifying glaucoma. We evaluate the proposed LSFMFT-RKDBN model on the fundus image database with qualitative and quantitative results analysis. The results are conducted by considering different metrics, such as peak signal to noise ratio (PSNR) and disease detection accuracy by using the different number of images.
Cite this Research Publication : P. M. Siva Raja, R. P Sumithra, K. Ramanan, Teager-Kaiser Boost Clustered Segmentation of retinal fundus images for Glaucoma Detection, Machine Learning and Deep Learning Techniques for Medical Science, CRC Press, Taylor & Francis Group, FL, USA, 2021.