Abstarct: Recognizing the license plate using OCR systems for a low-resolution image is a very challenging task. The use of high-quality cameras is a costly solution to this problem. In contrast, super-resolution techniques in image processing aim to provide a high-resolution image from single or multi-images of the scene. In this thesis, we propose a method for super-resolution of license plate given multi-frxames of the vehicle. For this purpose, in two steps the input frxames are registered on the reference image of the license plate. Then the reference image of the license plate is up-sampled. We propose a method to estimate the intensity of pixels with unknown value using weighted averaging in the neighboring pixels from video frxames. For evaluating the proposed method we used the recognition rate of an OCR system before and after super-resolution as well as SSIM criterion. The experimental result shows that applying the proposed method on low-resolution frxames improves 59.28% of the recognition rate for the license plate. Moreover, the SSIM criterion and visual assessment confirm the capability of the proposed method for super-resolution.