Malignancy or carcinoma is the unrestrained development of cells. There are numerous kinds of malignancies. In this, we are concentrating on one of the normal tumor of females is malignant growth of the cervix. Cervical cancer is the second most kind of malignant growth found in females, apart from the breast cancer which is first existence. There are abundant quantities of screening tests for Cervical cancer in which Pap smear has done. Pap smear is a virtuous device for first screening of this cancer yet it has constraints as there are consistently odds of blunder due to human perceptions. The main aim of this research is to preclude failures by utilizing computerized methods for identifying the cervical cell. Here, the image of Pap smear has been improved by utilizing Kaun Filter. The weight factor in kaun filter is resolved using an optimization technique called Bayesian Optimization Algorithm. Thus it tends to be an upgraded KF. The reformed picture has been segmented by Active Contour model. In this, the weight upgrade issue has been rectified using Analytic Hierarchy Process optimization technique. Hereafter, the solid features are eliminated from the segmented region which is most significant for identifying the cancer by utilizing ELMCSA (Extreme Learning machine with Crow Search optimization) classifier. Experimental results for the cervical cancer identification by the proposed ELM-CSA outperforms 93.5% of accuracy, 88.7% of specificity,79.21% of precision,92.25% of recall and 79.26% of F-measure than the existing classifiers such as ELM, ENN-TLBO, SVM and RBFN