摘要
针对车牌识别系统实时动态监控的需要,设计了一种基于Cortex-A53的便携式车牌识别系统。本系统以Cor-tex-A53架构的CPU为主控芯片,通过构建车辆图像能量图来寻找车牌位置,使用神经网络中BP算法进行车牌特征识别;同时,采用Qt平台开发软件系统界面。实验结果显示,该便携式车牌识别系统能够从不同角度、在不同距离及不同照明条件下对车牌图像进行识别,具有较强的抗干扰性和鲁棒性。
To meet the need of real-time and dynamic monitoring,a portable license plate recognition system based on Cortex-A53 is designed. The system uses Cortex-A53 as the core.By constructing energy image of the car,the system can locate the vehicle license plate location,and using the two-stage BP neural network,the license plate characters can be classified.Meanwhile,the Qt platform is used to design the software system interface.The experiment results show that the portable license plate recognition system can recognize the license plate images from different angles,different distances and different lighting conditions,and has strong anti-interference ability and robustness.
作者
李宇成
贺帅
Li Yucheng;He Shuai(College of Electrical and Control Engineering,North China University of Technology,Beijing 100144,China)
出处
《单片机与嵌入式系统应用》
2019年第6期79-82,共4页
Microcontrollers & Embedded Systems
