The combination of computer vision and deep learning approaches has changed automated systems across numerous domains. Such a domain is object detection. This study presents an automatic boom gate access method based on the YOLOv8 (You Only Look Once version 8) object detection model and license plate recognition (LPR) technology. It tries to resolve the issue of secure and efficient boom gate entry in restricted regions. The approach takes advantage of YOLOv8’s capacity to reliably detect and recognize license plates in real-time, allowing for automated gate operation. The experimental approach on the YOLOv8 framework used an open-sourced vehicle plate number dataset, to train and test the model, thereby reporting an average precision of 80%, a Mean Average Precision (mAP) of 76% at a threshold of 50%, and an average F-1 score of 75%. Experimental tests with live samples demonstrated how well and consistently the proposed technique performs in a variety of environmental situations, including lighting, weather, and vehicle orientations. The findings demonstrated the YOLOv8-based system’s effective application in real-world circumstances.