License plate detection and recognition are critical components of intelligent transportation systems, supporting applications such as automated toll collection, parking management, and traffic monitoring. This study introduces a robust system specifically designed for detecting and recognizing Bangla vehicle license plates. Leveraging YOLOv8, a state-of-the-art object detection model, and EasyOCR for optical character recognition, the system effectively addresses the complexities of Bangla scripts. A custom dataset was created, consisting of 276 training, 79 validation, and 38 testing images. These images were meticulously annotated using Roboflow to ensure high-quality training data. The system demonstrated impressive performance, achieving detection accuracy between 93% and 96%, a Character Error Rate (CER) of 1.52%, and a Word Error Rate (WER) of 5.53%. These results underscore its capability to handle challenges such as skewed plates, partial occlusions, and varying lighting conditions. The training process was validated through performance curves showing steady improvements in loss and accuracy, while visualizations of the detection pipeline further highlighted the system's ability to accurately localize and process license plates in diverse environments. Despite its success, the system faces challenges with noisy inputs, low-resolution images, and non-standardized fonts. Future work will focus on addressing these limitations to enhance scalability and robustness. This study provides a reliable, scalable solution for Bangla license plate detection and recognition, advancing intelligent transportation in Bangladesh