Cervical cancer remains one of the leading causes of mortality among women worldwide, and its early detection is crucial to improve survival rates. While a Pap smear is widely used as a diagnostic tool, it has limitations in sensitivity and specificity due to the inherent subjectivity of cytological analysis. This study proposes a methodology for cervical cell segmentation and extraction based on the Laplacian of Gaussian (LoG) algorithm, which enables the generation of regions of interest to detect and segment cells precisely in cervical cytology samples. Over 2,000 digital images of Pap smear slides were analyzed, derived from 500 cervical cytology slides provided by the State Public Health Laboratory of Michoacán, México. The dataset results demonstrated an accuracy of 96.5%, a recall rate of 99.2%, and an F-measure of 97.8%. Furthermore, the methodology was optimized for real-time analysis, allowing efficient segmentation and detection of cells and their morphological variations. This methodology not only significantly improves accuracy and efficiency in cervical cell segmentation but also has a high potential for application in other experiments that require precise cell segmentation despite morphological variations. In this regard, it offers an adaptable and versatile approach, making a substantial contribution to cytological studies and establishing itself as an effective process to extract cervical cells automatically in real time.