Globally, cervical cancer is one of the top causes of death for women, especially in areas with poor access to medical treatment. Early detection and diagnosis are critical for effective treatment, yet conventional screening techniques, such HPV testing and Pap smears, often face challenges related to subjectivity, time constraints, and the need for skilled professionals. To overcome these challenges, a novel deep learning-based ACERNET model is proposed an automated cervical cancer detection. Initially, the input cervicography images are collected from Publicly available dataset. Then the image is preprocessed using Gaussian filters (GF) to enhance clarity by reducing noise. The deep learning-based ResNet50 is used for extracting features from the Cervical cancer. Finally, the extracted features are subsequently fed into Random Forest (RM) classifier that classify Adenocarcinoma and Squamous cell carcinoma. The effectiveness of the proposed ACERNET was evaluated using F1 score, accuracy, precision, recall, and specificity. The proposed ACERNET model achieved a classification accuracy of 98.88%. The proposed model enhanced the total accuracy by 0.67%, 5.91%, 1.04% better than Cervi Former, CYENET, C3Net respectively.