A Robust Layout-Independent License Plate Detection and Recognition Model Based on Attention Method

Auto License Plate Detection and Recognition (ALPDR) is required in many industrial fields, including visual surveillance systems vehicle registration control. Even though this field has recently demonstrated high performance according to rapid advances Deep Learning (DL) based technologies, there are still two problems that most related works have not yet solved. One layout-dependent the other problem of recognition accuracy degradation due unconstrained environment conditions. In paper, we present a more accurate, flexible layout-independent method improve LP detection under various outdoor For this, proposed new framework with lightweight efficient anchor-free networks, employing idea CenterNet attention-based networks residual deformable block suitable for LPR tasks. Various experiments demonstrate method, using famous benchmark datasets such as CCPD, AOLP VBLPD. The outperformed conventional methods. Additionally, Korean handicapped parking card (KHPC) data was tested prove usefulness marked character different license plates.