The tiny tissues that make up the retina are responsible for absorbing light and converting it into visual recognition in the brain. These people may experience vision loss and blindness as a result of damage to this vital organ. Thus, early diagnosis of the disease may prevent total blindness and, in certain situations, allow vision to return to normal. Therefore, timely and precise CSR detection prevents serious macular damage and is a foundation for detecting other retinal diseases. The OCT pictures have been used to identify CSR, but developing a system that is both accurate and computationally economical is still a difficult task. This research designs a framework for precise and automatically segmenting CSR from denoised OCT images using the Level set segmentation technique. It helps to identify the central serous retinopathy from the OCT images. The recommended system is assessed using the method of the healthy macula and central serous retinopathy OCT images. The proposed system provides an assessment based on the Dice Coefficient (DC), Jaccard Index(JC), and HD(Hausdorff Distance). The outcome of this Level-set segmentation technique compared with the Graph-based segmentation approach. Among these two techniques Level set segmentation produces better results in terms of DC, JC, and HD.