The urge to unveil the secrets of digital visual enhancement has always been a dream for mankind. It has always been an expanding realm of research that has never failed to surprise humanity. In this paper, we have proposed a modified Clustering technique in Fuzzy C-Means named Narrow Fuzzy C-Means Clustering. This clustering method is implemented and fused with U-Net Convolution. The proposed segmentation algorithm uses this unique technique which assists in providing elevated and enhanced outcomes. The suggested approach helps to precisely segment the area of interest from the provided input images. The novel proposal provides an accuracy of 96.5% with a Dice Similarity Co-Efficient (DSC) of 0.94 which tends to determine the exact segmentation of the area of interest with a low false positive rate.