Cervical cancer, a disease that affects women the most. When a woman's cervix undergoes some modifications, this is visible. These cancer cells have the ability to spread to another parts, including bladder, livers, lungs, rectum, further complicating situations. Higher rates of recovery have been reported with earlier discovery, screening, and rigorous procedures. This paper proposes a novel technique to detect cervical cancer utilizing deep learning techniques. The idea is to provide improvement in the efficiency, the network used to cluster the images using Optimal Kernelized Fuzzy C-Means (OKFCM) clustering along with classified image to give high-resolution image classification.Then to build an accurate cervical cancer histopathological image ensemble based feature extraction model, which is the very first and most important procedure in our system. An artificial neural network is used to distinguish between normal, abnormaland malignant cells, producing accurate findings than manual testing procedures like Pap test and LBC (Liquid Based cytology). Database of 962histopathological cervicalimages used to test our technique.The outcomes of all trials proved that this technique produced the best results inPrecision, Specificity,Recall, AUC,Accuracy, FPR and FNR. This strategy can help pathologists with cervical disease categorization by reducing their cognitive load and increasing their diagnostic efficiency and accuracy. It will have the ability to be used in medical care for the diagnosis of cervical cancer histology.