The condition of the human retina is essential for detecting eye conditions that can cause blindness if left untreated, such as Choroidal Neovascularization (CNV), Diabetic Macular Edema (DME) and Drusen. By clearly displaying the anatomy of the retina, the analysis of retinal layers in Optical Coherence Tomography (OCT) images improves diagnostic precision. This study introduces an automated approach that uses deep learning and sophisticated image processing techniques to precisely separate retinal layers. The system incorporates segmentation models to precisely delineate the retinal layers, feature extraction to highlight significant retinal features, and image enhancing approaches to increase quality. In order to provide dependable and consistent results, it tackles challenges such noise, poor contrast OCT pictures, and inconsistencies present in manual segmentation. The approach utilizes Convolutional Neural Networks (CNNs) with the U-Net architecture, which is tailored for medical image analysis, ensuring quick and precise segmentation of even subtle retinal details.