The article analyzes the impact of a vehicle’s license plate surroundings on the accuracy of low-resolution image registration in super-resolution algorithms. By utilizing these surroundings, it is possible to correctly recognize vehicle license plates in motion, even when they are illegible in individual frames of a video sequence. The presented experimental results demonstrate the usefulness of additional information visible in the immediate vicinity of the plate (e.g., radiator, bumper, headlights) in the registration process, which allows for obtaining a resulting image of significantly higher quality and better readability than in the case of using only low-resolution license plate images.Automatic recognition of vehicle license plates in motion is one of the most important software elements of Intelligent Transport Systems. (…)
The article contains 17834 characters.
Source: Logistyka Magazine








