|
|
论文:2021,Vol:39,Issue(4):891-900 |
|
|
引用本文： |
|
|
尉天成, 陈小锋, 殷元亮. 基于多尺度卷积神经网络的道路交通标志识别方法研究[J]. 西北工业大学学报 |
|
|
WEI Tiancheng, CHEN Xiaofeng, YIN Yuanliang. Research on traffic sign recognition method based on multi-scale convolution neural network[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 基于多尺度卷积神经网络的道路交通标志识别方法研究 |
|
| 尉天成, 陈小锋, 殷元亮 |
|
| 西北工业大学 自动化学院, 陕西 西安 710129 |
| 摘要: |
| 为了准确识别不同路况下的交通标志信息，提出一种在Faster RCNN模型基础上改进的深度学习方法。针对交通标志的显著特征，设计了多路并联全卷积神经网络，对原始图像中的交通标志颜色、形状以及纹理进行多路特征提取，将多路特征层进行融合得到最终特征图，通过图像预处理加强了模型在多种环境和天气状况下的适应能力。同时在特征提取网络中加入深浅层特征层的融合特征，保留浅层特征层的细节纹理信息和深层特征层的语义信息，得到最终特征层能够适应多尺度变化的交通标志的识别。在原有RPN候选区域生成网络前，利用交通标志先验知识作为辅助进行目标检测定位，提出了针对交通标志识别更加合理的候选锚框生成办法。从先验知识统计交通标志尺寸和比例结果出发，设计适用于交通标志识别的目标候选框，减少了大量冗余的和负相关的候选框，提高检测准确度减少检测时间；加入针对深浅特征层的多尺度候选框生成方法，在强化多尺度目标识别能力的同时，进一步加强了小目标检测和识别效果；采用国际通用交通标志规范数据集GTSRB/GTSDB以及国内交通标志数据集TT100K对模型识别能力进行识别验证。 |
| 关键词:
交通标志识别
深度学习
卷积神经网络
特征融合
小目标识别
|
|
| Research on traffic sign recognition method based on multi-scale convolution neural network |
|
| WEI Tiancheng, CHEN Xiaofeng, YIN Yuanliang |
|
| School of Automation, Northwestern Polytechnical University, Xi'an 710129, China |
| Abstract: |
| In order to accurately identify the traffic sign information under different road conditions, an improved deep learning method based on Faster RCNN model is proposed. Firstly, a multi-channel parallel full convolution neural network is designed to extract the color, shape and texture features of traffic signs in the original image. The multi-channel feature layers are fused to get the final feature map, and the adaptability of the model in various environment and weather conditions is enhanced by the image preprocessing. At the same time, the fusion features of deep and shallow feature layer are added into the feature extraction network, and the detailed texture information of shallow feature layer and semantic information of deep feature layer are retained, and the final feature layer can adapt to multi-scale change of traffic sign recognition. Secondly, the prior knowledge of traffic signs is used to detect and locate the target before the original RPN candidate region is generated. A more reasonable method for generating feature points and candidate anchor frames for traffic sign recognition is proposed. Based on the prior knowledge statistics of traffic sign size and proportion results, a target candidate frame suitable for traffic sign recognition is designed, a large number of redundant and negative correlation candidate frames is reduced, the detection accuracy and reduces the detection time is improved; secondly, the multi-scale candidate frame generation method for the deep and shallow feature layer is added to enhance the multi-scale target recognition ability and further strengthen the multi-scale target recognition ability Finally, this paper uses the international general traffic sign specification data set GTSRB/GTSDB and domestic traffic sign data set tt100k to verify the recognition ability of the model. |
| Key words:
traffic sign recognition
deep learning
convolutional neural network
feature fusion
small target recognition
|
|
| 收稿日期: 2020-11-23
修回日期:
|
| DOI: 10.1051/jnwpu/20213940891 |
| 基金项目: 陕西省重点研发计划（2019GY-117）资助 |
| 通讯作者: 陈小锋(1974-),西北工业大学副教授,主要从事智能交通系统研究。e-mail:chenxf@nwpu.edu.cn
Email：chenxf@nwpu.edu.cn |
| 作者简介: 尉天成(1998-),西北工业大学硕士研究生,主要从事交通图像分析研究。
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] ZHANG J, WANG F, WANG K, et al. Data-driven intelligent transportation systems:a survey[J]. IEEE Trans on Intelligent Transpor-tation Systems, 2011, 12(4):1624-1639
[2] SEO Y, LEE J, ZHANG W, et al. Recognition of highway workzones for reliable autonomous driving[J]. IEEE Trans on Intelligent Transportation Systems, 2015, 16(2):708-718
[3] FAN Yanjun, ZHANG Weigong. Traffic sign detection and classification for advanced driver assistant systems[C]//International Conference on Fuzzy Systems and Knowledge Discovery(FSKD), Zhangjiajie, 2015:1335-1339
[4] 戈侠,于凤芹,陈莹. 基于分块自适应融合特征的交通标志识别[J]. 计算机工程与应用,2017,53(3):188-192 GE Xia, YU Fengqin, CHEN Ying. Traffic sign recognition based on blocking self-adaptation fusion features[J]. Computer Engineering and Applications, 2017, 53(3):188-192(in Chinese)
[5] 沙莎,肖学钢. 一种多通道融合的交通标志检测方法[J]. 计算机工程,2009,35(6):202-204 SHA Sha, XIAO Xuegang. Traffic sign detection method with multi-channel integration[J]. Computer Engineering, 2009, 35(6):202-204(in Chinese)
[6] MADANI A, YUSOF R. Traffic sign recognition based on color, shape, and pictogram classification using support vector machines[J]. Neural Computing & Applications, 2017, 36(5):1-11
[7] AZIZ S, MOHAMED E A, YOUSSEF F. Traffic sign recognition based on multi-feature fusion and ELM classifier[J]. Procedia Computer Science, 2018, 127(1):146-153
[8] BOUMEDIENE M, CUDEL C, BASSET M, et al. Triangular traffic signs detection based on RSLD algorithm[J]. Machine Vision and Applications, 2013, 24(8):1721-1732
[9] LILLO-CASTELLANO J M, JIMÉNEZ I, POZUELO C, et al. Traffic sign segmentation and classify-cation using statistical learning methods[J]. Neurocomputing, 2015, 153:286-299
[10] GIRSHICK R. Fast R-CNN[J/OL]. (2015-09-27)[2020-09-01]. https://arxiv.org/pdf/1504.08083.pdf
[11] CHENG P, LIU W, ZHANG Y, et al. LOCO:local context based faster R-CNN for small traffic sign detection[C]//International Conference on Multimedia Modeling, Cham, 2018
[12] XIONG C, CONG W, MA W, et al. A traffic sign detection algorithm based on deep convolutional neural network[C]//IEEE International Conference on Signal and Image Processing, 2016
[13] REN Shaoqing, HE Kaiming, GIRSHICK Ross, et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149
[14] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Trans on Patcern Analysis and Machrne Interlligence, 2015, 39(4):640-651
[15] 李凯,韩冰,张景滔. 基于条件随机场与多尺度卷积神经网络的交通标志检测与识别[J]. 计算机应用, 2018,38(增刊2):270-275 LI Kai, HAN Bing, ZHANG Jingtao. Traffic sign detection and recognition based on oonditional random field and multi-scale convolutional neural network[J]. Journal of Computer Applications, 2018,38(suppl 2):270-275(in Chinese)
[16] HOUBEN S, STALLKAMP J, SALMEN J, et al. Detection of traffic signs in real-world images:the German traffic sign detection benchmark[C]//The 2013 International Joint Conference on Neural Networks, Dallas, TX, 2013:1-8
[17] ZHU Z, LIANG D, ZHANG S, et al. Traffic-sign detection and classification in the wild[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition, 2016 |
|
|
|
相关文献: |
|
|
1．张科, 苏雨, 王靖宇, 王霰宇, 张彦华.基于融合特征以及卷积神经网络的环境声音分类系统研究[J]. 西北工业大学学报, 2020,38(1): 162-169 |
|
|
2．李林, 张盛兵, 吴鹃.面向图像识别的深度学习VLIW处理器设计[J]. 西北工业大学学报, 2020,38(1): 216-224 |
|
|
3．张飞, 闫杰.基于深度卷积神经网络的云分类算法[J]. 西北工业大学学报, 2020,38(4): 740-746 |
|
|
4．白芃远, 许华, 孙莉.基于卷积神经网络与时频图纹理信息的信号调制方式分类方法[J]. 西北工业大学学报, 2019,37(4): 816-823 |
|
|
|
|
|
|
