Computer vision coupled with Deep Learning (DL) techniques bring out a substantial prospect in the field of traffic control, monitoring and law enforcing activities. This paper presents a YOLOv4 object detection model in which the Convolutional Neural Network (CNN) is trained and tuned for detecting the license plate of the vehicles of Bangladesh and recognizing characters using tesseract from the detected license plates. Here we also present a Graphical User Interface (GUI) based on Tkinter, a python package. The license plate detection model is trained with mean average precision (mAP) of 90.50% and performed in a single TESLA T4 GPU with an average of 14 frames per second (fps) on real time video footage.