Automatic license plate recognition (ALPR) has become a crucial technology in various applications such as traffic management, law enforcement, and access control systems. This thesis presents an advanced ALPR system leveraging Convolutional Neural Networks (CNNs), HAAR cascade classifiers, and Optical Character Recognition (OCR) to achieve high accuracy in license plate detection and character recognition. The proposed system consists of three primary stages: license plate detection using HAAR cascades, character segmentation, and character recognition using CNNs integrated with OCR techniques. In the first stage, the HAAR cascade classifier efficiently detects license plates in diverse and challenging conditions, including varying lighting and weather scenarios. The second stage involves segmenting the detected license plates into individual characters, which are then processed for recognition. In the final stage, a CNN model is employed to accurately recognize segmented characters, leveraging OCR to refine and verify the results. Extensive experiments were conducted on various datasets to evaluate the performance of the proposed system. The results demonstrate that our approach achieves superior accuracy and robustness compared to traditional methods, particularly in complex environments. This research contributes to the advancement of ALPR technology, providing a reliable and efficient solution for real-world applications