License plate detection and recognition based on deep learning

With the prevalence of new energy vehicles, the demand for efficient detection and recognition of their license plates has greatly increased. To address this issue, this paper divides license plate detection and recognition into two parts. Firstly, the paper utilizes a target detection algorithm in deep learning (YOLOv3-spp) to accomplish the task of license plate detection and positioning. Then, the LPRnet network is used to recognize license plate characters. Finally, the two are integrated to form a network model for the detection and recognition of license plates on new energy vehicles. The results indicate that the model performs at a relatively high level.