The article presents the impact of the rotation angle of a camera recording an image sequence on the readability of license plate images obtained using a super-resolution algorithm. The resulting image is subjected to subsequent phases of the recognition process, the outcome of which is largely dependent on the quality of the input image, especially in the case of recording low-resolution video sequences.License plates are currently the primary method of vehicle recognition, allowing for vehicle tracking within Intelligent Transportation Systems (ITS) [3,11] and linking to a specific owner, which is significant for both the police and safety improvement systems. A future-oriented technique, slowly being implemented in the field of automated toll collection, e.g., for highway travel, is RFID (Radio Frequency IDentification), however, the use of license plate images remains dominant. One of the reasons for this state of affairs is the fact that a license plate number can be read without the use of specialized equipment, as is the case with RFID. (…)
The article contains 14023 characters.
Source: Logistyka Magazine