<![CDATA[The rising number of motor vehicles has created a need for rapid and accurate vehicle identification systems. License Plate Recognition (LPR) systems are widely used in traffic monitoring, parking management, access control, and security. However, their performance is influenced by factors such as lighting, capture angles, image quality, and the condition of the license plates. This study employs a combination of OpenCV and deep learning-based Optical Character Recognition (OCR) to detect and recognize license plate characters. OpenCV is utilized for image preprocessing and license plate localization, while OCR is used for character recognition. Testing was conducted on 100 vehicle images featuring variations in lighting, positioning, and license plate quality. The results demonstrate that optimization through image preprocessing, data augmentation, and model fine-tuning increased accuracy from 78.6% to 85.2%. Processing time also decreased from 0.46 to 0.35 seconds per image, thereby supporting real-time applications. The integration of OpenCV and deep learning-based OCR proved effective in enhancing the accuracy and efficiency of automated vehicle identification systems, supporting intelligent transportation and computer vision-based security.]]>
Copyrights © 2026
