摘要
交通标志识别是实现智能驾驶以及无人驾驶的关键技术之一,已成为智能交通领域的研究热点。深度学习具有强大的图像识别能力,因此将深度学习技术中的卷积神经网络算法应用于交通标志识别。在基本的卷积网络结构中增加了新的“压平层”与“丢弃层”,提高了训练能力且节省时间。实验结果表明,能将识别率提升到95%以上,具有自动学习的能力和训练周期短的优点,并且准确性高,鲁棒性好,具有良好的泛化能力。
Traffic sign recognition is one of the key technologies to realize intelligent driving and driverless driving,which has become a research hotspot in the field of intelligent transportation.Deep learning has a strong ability of image recognition,so convolution neural network algorithm of deep learning technology is applied to traffic sign recognition.A new"flattening layer"and"discarding layer"are added to the basic convolution network structure to improve the training ability and save time.The experimental results show that it can improve the recognition rate to more than 95%,has the advantages of automatic learning and short training cycle,high accuracy,good robustness and good generalization ability.
作者
李金樽
罗山
Li Jinzun;Luo Shan(School of Traffic and Automotive Engineering,Panzhihua University,Panzhihua Sichuan 617000,China)
出处
《山西电子技术》
2020年第5期21-23,共3页
Shanxi Electronic Technology
基金
攀枝花学院大学生创新创业训练计划项目(2019cxcy023)。
关键词
深度学习
卷积神经网络
交通标志识别
deep learning
convolution neural network
traffic sign recognition