This study presents an Automatic License Plate Recognition (ALPR) system implemented on Raspberry Pi 5, a low-cost edge computing device. The system employs the YOLO v11 object detection model for license plate detection, achieving a mean Average Precision (mAP) of 0.995 at an IoU threshold of 0.5 on the UFPR-ALPR dataset. To improve license plate recognition accuracy, the system leverages a robust recognition model based on MobileViTV2 and is applied to the license plate region detected from a single input image. Evaluated on the UFPR-ALPR dataset, the overall system achieves an accuracy of 98.33%. Additionally, CPU optimizations were implemented to ensure efficient inference on the Raspberry Pi 5's limited computational resources. These results highlight the potential for deploying highly accurate ALPR systems on edge devices, serving critical applications such as security and traffic monitoring. The source code for this project is publicly available and can be accessed on GitHub.