首页 > 论文 > 激光与光电子学进展 > 55卷 > 12期(pp:121009--1)

基于改进深度卷积神经网络的交通标志识别

Traffic Sign Recognition Based on Improved Deep Convolution Neural Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

在实际交通环境中, 所采集到的交通标志图像质量往往受到运动模糊、背景干扰、天气条件以及拍摄视角等因素的影响, 这对交通标志自动识别的准确性、实时性和稳健性提出了很大的挑战。为此提出了改进深度卷积神经网络AlexNet的分类识别算法模型, 该模型在传统AlexNet模型基础上, 以真实场景中拍摄的交通标志图像数据集GTSRB为研究对象, 将所有卷积层的卷积核修改为3×3大小, 为了预防和减少过拟合的出现在两个全连接层后加入dropout层, 并且为了提高交通标志识别精度, 在网络模型第5层后增加两层卷积层。实验结果表明, 改进后AlexNet模型在交通标志识别方面具有一定的先进性和稳健性。

Abstract

In the actual traffic environment, the quality of the collected traffic signs is often influenced by the factors such as motion blur, background interference, weather conditions and shooting angles and so on, which poses a great challenge to the accuracy, real-time and robustness of traffic sign automatic identification. Owing to this, a classification recognition algorithm model of improved deep convolution neural network AlexNet is proposed. On the basis of the traditional AlexNet model, this model takes the traffic sign image data set GTSRB taken in the real scene as the research object, modifies the convolution kernels of all coiling layers to 3×3, in order to prevent and reduce the occurrence of over fitting, the dropout layer is added after two fully connected layers. In order to improve the accuracy of traffic sign recognition, two convolution layers are added after the fifth layer of the network model. The experimental results show that the improved AlexNet model is advanced and robust in traffic sign recognition.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:TP391.7

DOI:10.3788/LOP55.121009

所属栏目:图像处理

基金项目:国家自然科学基金(41461078)

收稿日期:2018-04-25

修改稿日期:2018-05-28

网络出版日期:2018-07-12

作者单位    点击查看

马永杰:西北师范大学物理与电子工程学院, 甘肃 兰州 730070
李雪燕:西北师范大学物理与电子工程学院, 甘肃 兰州 730070
宋晓凤:西北师范大学物理与电子工程学院, 甘肃 兰州 730070

联系人作者:马永杰(myjmyj@163.com)

【1】Xu Y, Wei Z Y. An improved traffic sign image recognition algorithm[J]. Laser & Optoelectronics Progress, 2017, 54(2): 021001.
徐岩, 韦镇余. 一种改进的交通标志图像识别算法[J]. 激光与光电子学进展, 2017, 54(2): 021001.

【2】Zhang S F, Zhu T. A method of traffic sign detection and recognition based on HDR technology[J]. Laser & Optoelectronics Progress, 2018, 55(9): 091006.
张淑芳, 朱彤. 一种基于HDR技术的交通标志牌检测和识别方法[J]. 激光与光电子学展, 2018, 55(9): 091006.

【3】Luo H L, Yang Y, Tong B, et al. Traffic sign recognition using a multi-task convolutional neural network[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 19(4): 1100-1111.

【4】Zhong X M, Yu G Z, Ma Y L, et al. Research on traffic sign recognition algorithms based on fast regional convolution neural network[C]∥Annual Conference Papers of China Society of Automotive Engineers, 2016: 4.
钟晓明, 余贵珍, 马亚龙,等.基于快速区域卷积神经网络的交通标志识别算法研究[C]∥中国汽车工程学会年会论文集, 2016: 4.

【5】Tan T Z, Lu J B, Wen J W, et al. Traffic sign recognition applying with convolution neural network and RPN[J]. Computer Engineering and Applications, 2018, 54 (21): 251-256.
谭台哲, 卢剑彪, 温捷文, 等. 应用卷积神经网络与RPN的交通标志识别[J]. 计算机工程与应用, 2018, 54 (21): 251-256.

【6】Xu Z, Feng C H.Modified scale dependent pooling model for traffic image recognition[J]. Journal of Computer Applications, 2018, 38(3): 671-676.
徐喆, 冯长华. 用于交通图像识别的改进尺度依赖池化模型[J]. 计算机应用, 2018, 38(3): 671-676.

【7】Dang Y, Zhang J X, Deng K Z, et al. Study on the evaluation of land cover classification using remote sensing images based on AlexNet[J]. Journal of Geo-Information Science, 2017, 19(11): 1530-1537.
党宇, 张继贤, 邓喀中, 等. 基于深度学习AlexNet的遥感影像地表覆盖分类评价研究[J]. 地球信息科学学报, 2017, 19(11): 1530-1537.

【8】Lü H M, Zhao D, Chi X B. Deep learning for early diagnosis of Alzheimer''s disease based on intensive AlexNet[J]. Computer Science, 2017, 44(6A): 50-60.
吕鸿蒙, 赵地, 迟学斌. 基于增强AlexNet的深度学习的阿尔茨海默病的早期诊断[J]. 计算机科学, 2017, 44(6A): 50-60.

【9】Zhang Y M, Chang F L, Li N J, et al. Modified AlexNet for dense crowd counting[C]∥2017 2nd International Conference on Computer Science and Engineering, Information Science and Internet Technology (CII 2017): 351-357.

【10】Chen Q J, Li Y, Chai Y Z. A multi focus image fusion algorithm based on deep learning[J]. Laser & Optoelectronics Progress, 2018, 55(7): 071015.
陈清江, 李毅, 柴昱洲. 一种基于深度学习的多聚焦图像融合算法[J]. 激光与光电子学进展, 2018, 55(7): 071015.

【11】Li J N, Zhang B H. Face recognition by feature matching fusion combined with improved convolutional neural network[J]. Laser & Optoelectronics Progress, 2018, 55(10): 101504.
李佳妮, 张宝华. 特征匹配融合结合改进卷积神经网络的人脸识别[J]. 激光与光电子学进展, 2018, 55(10): 101504.

【12】Xiao J S, Liu E Y, Zhu L, et al. Improved image super-resolution algorithm based on convolutional neural network[J]. Acta Optica Sinica, 2017, 37(3): 0318011.
肖进胜, 刘恩雨, 朱力, 等. 改进的基于卷积神经网络的图像超分辨率算法[J]. 光学学报, 2017, 37(3): 0318011.

【13】Hinton G E, Srivastava N, Krizhevsky A, et al. Improving neural networks by preventing co-adaptation of feature detectors[J]. Computer Science, 2012, 3(4): 212-223.

【14】Stallkamp J, Schlipsing M, Salmen J, et al. Man vs. computer: Benchmarking machine learning algorithms for traffic sign recognition[J]. Neural Networks, 2012, 32: 323-332.

【15】Liu Z W, Zhao X M, Li Q, et al. Traffic sign recognition method based on graphical model and convolutional neural network[J]. Journal of Traffic and Transportation Engineering, 2016, 16(5): 122-131.
刘占文, 赵祥模, 李强, 等. 基于图模型与卷积神经网络的交通标志识别方法[J]. 交通运输工程学报, 2016, 16(5): 122-131.

【16】Ou X F, Xiang C Q, Guo L Y, et al. Research of recognition of digital characters on vehicle license based on caffe deep learning framework[J]. Journal of Sichuan University(Natural Science Edition), 2017, 54(5): 971-977.
欧先锋, 向灿群, 郭龙源, 等. 基于Caffe深度学习框架的车牌数字字符识别算法研究[J]. 四川大学学报(自然科学版), 2017, 54(5): 971-977.

【17】Ruta A, Li Y M, Liu X H. Real-time traffic sign recognition from video by class-specific discriminative features[J]. Pattern Recognition, 2010, 43(1): 416-430.

引用该论文

Ma Yongjie,Li Xueyan,Song Xiaofeng. Traffic Sign Recognition Based on Improved Deep Convolution Neural Network[J]. Laser & Optoelectronics Progress, 2018, 55(12): 121009

马永杰,李雪燕,宋晓凤. 基于改进深度卷积神经网络的交通标志识别[J]. 激光与光电子学进展, 2018, 55(12): 121009

被引情况

【1】徐健,吴曙培,刘秀平. 基于改进深度神经网络的纱管分类. 激光与光电子学进展, 2020, 57(10): 101014--1

【2】童英,杨会成. 基于改进神经网络的交通标志识别. 激光与光电子学进展, 2019, 56(19): 191002--1

【3】刘力荣,唐新明,赵文吉,高小明,谢俊峰. 基于影像与激光数据的小交标检测与地理定位. 中国激光, 2020, 47(9): 910002--1

【4】张蕾,蔡明. 融合卷积神经网络与主题模型的图像标注. 激光与光电子学进展, 2019, 56(20): 201004--1

【5】郭进祥,刘立波,徐峰,郑斌. 基于YOLO v3的机场场面飞机检测方法. 激光与光电子学进展, 2019, 56(19): 191003--1

【6】吴沛佶,梅雪,何毅,袁申强. 基于深度网络模型的视频序列中异常行为的检测方法. 激光与光电子学进展, 2019, 56(13): 131101--1

【7】吉祥凌,吴军,易见兵,张晓光. 基于深度学习的管制物品自动检测算法研究. 激光与光电子学进展, 2019, 56(18): 180402--1

【8】梁令羽,张天天,何为. 多尺度卷积神经网络的头部姿态估计. 激光与光电子学进展, 2019, 56(13): 131003--1

【9】胡敬锋,张秀再,杨昌军. 基于改进M型卷积网络的RGB彩色遥感图像云检测. 激光与光电子学进展, 2019, 56(16): 162804--1

【10】白睿,徐友春,李永乐,李炯,谢枫. 智能车道路场景数字字符识别技术. 激光与光电子学进展, 2019, 56(19): 191506--1

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF