Human intrusion detection is a critical concern in the field of security, especially for finding illicit entrance at private or restricted places. Unauthorized intrusions occur when attackers gain illicit access to residential or restricted areas and engage in property theft. Security monitoring is essential in various locations, including residential areas and households, particularly during specified periods, typically from 10 pm to 6 am. In this study, we propose an approach for detecting human intrusion with YOLOv8 for security cameras. The model was identified and tracked objects in the predetermined ROI area, and got over 80% accuracy from human identification data. After training the model, the software was employed with several fundamental functions for detection and tracking of unauthorized human intrusions.