This proposed research investigates how computer vision methods and artificial intelligence may improve cervical cancer screening and diagnosis, particularly relevant in settings with limited resources. Two different AI-based methods for cervical image categorization have been explored using the IARC databank and off-the-shelf dataset. Developed an algorithm to preprocess the images—removing specular reflections and segmenting the cervix. Preliminary results indicate a statistically significant improvement in accuracy by using a novel two-stage model approach giving 95% accuracy for the differentiating between suspicious, and non-suspicious images. These results suggest that applying sophisticated computational methods and AI-driven approaches may improve the accuracy and efficiency of cervical cancer screening helping medical practitioners to potentially minimize diagnostic errors, and enhance early detection rates having particular relevance in low- and middle-income countries (LMICs) where traditional screening methods face significant obstacles.