OCT has completely changed ocular imaging by providing fine-grained cross-sectional views of the retinal layers. However, because to noise, inconsistent image quality, and pathological characteristics, it is still difficult to accurately segment retinal borders in OCT images. This study provides a thorough approach to improve retinal boundary segmentation accuracy andconsistency. The method comprises layer-wise color division representation postprocessing, thresholding with edge detection, enhanced filtering, database collection, and performance assessment. To ensure adaptation across patient demographics and eye diseases, a broad OCT dataset is obtained. Preprocessing removes noise and artifacts from images by standardizing their format and resolution. Visibility is improved by enhanced filtering using Gaussian filters and adaptive histogram processing, which is followed by Sobel edge detection and Kapur thresholding. Mean thresholding and interpolation approaches are examples of postprocessing. Interpretability of the results is improved by layer-wise color splitting. Through cross-validation and hand annotations, performance is thoroughly assessed. This work advances ophthalmic image analysis by providing a thorough method for segmenting the retinal boundaries in OCT pictures.