基于CNN和Bagging集成的交通标志识别

doi:10.16088/j.issn.1001-6600.2021102203

摘要/Abstract

摘要： 针对直接集成简单分类器对交通标志数据库进行识别出现的类别预测效果较差的问题,提出一种基于卷积神经网络(CNN)和Bagging集成学习的交通标志识别算法,采用爬虫和图像增强技术实现交通标志数据集的扩充,以CNN网络提取交通标志图像的特征,通过采用最大池化层实现图像数据下采样,采用较浅的网络深度以简化整体网络结构。在CNN网络特征提取的基础上,利用软投票机制对多项Logistic、K近邻、SVM个体学习器进行集成,实现较准确的交通标志识别。实验结果表明,该算法在TSRD交通标志识别数据库测试集上的识别准确率达到了93.00%,相对于未改进的卷积神经网络模型识别准确率提高了11.99个百分点,并较高于通过VGG16和ResNet50迁移学习实现的识别准确率,具有较快的收敛速度。

关键词: 图像增强, 卷积神经网络, 迁移学习, 集成学习, 交通标志识别

Abstract: To solve the problem of poor category prediction effect of direct integration of simple classifiers to recognize traffic sign databases, a traffic sign recognition algorithm based on Convolutional Neural Network and Bagging ensemble learning is proposed, using crawler and image enhancement technology to expand the data set of traffic signs, and the features of traffic sign images is extracted by Convolutional Neural Network. For the characteristics of the logo image, the image data are down-sampled by using the maximum pooling layer, and the overall network structure is simplified by using a shallower network depth. On the basis of Convolutional Neural Network feature extraction, the soft voting mechanism is used to integrate Multiple Logistic, K nearest Neighbor, and SVM individual learners to obtain a better prediction result and achieve more accurate traffic sign recognition. Experimental results show that the recognition accuracy of the algorithm on the TSRD traffic sign recognition database test set has reached 93.00%, which is 11.99% higher than the recognition accuracy of the original Network model, and is higher than the recognition accuracy achieved by VGG16 and ResNet50 migration learning having a faster convergence speed.

Key words: image enhancement, convolutional neural network, transfer learning, integrated learning, traffic sign recognition

中图分类号:

U463.6

田晟, 宋霖. 基于CNN和Bagging集成的交通标志识别[J]. 广西师范大学学报（自然科学版）, 2022, 40(4): 35-46.

TIAN Sheng, SONG Lin. Traffic Sign Recognition Based on CNN and Bagging Integration[J]. Journal of Guangxi Normal University(Natural Science Edition), 2022, 40(4): 35-46.

参考文献

[1] AlAM A, JAFFERY Z A. Indian traffic sign detection and recognition[J]. International Journal of Intelligent Transportation Systems Research, 2020, 18(1):98-112. DOI: 10.1007/s13177-019-00178-1.
[2]刘亚辰. 交通标志识别系统算法研究与设计[D]. 武汉:武汉理工大学,2017.
[3]ARDIANTO S, CHEN C J, HANG H M. Real-time traffic sign recognition using color segmentation and SVM[C]// 2017 International Conference on Systems, Signals and Image Processing (IWSSIP).Piscataway, NJ: IEEE, 2017: 1-5. DOI: 10.1109/IWSSIP.2017.7965570.
[4]宗鸣,龚永红,文国秋,等.基于稀疏学习的KNN分类[J].广西师范大学学报(自然科学版),2016,34(3):39-45. DOI: 10.16088/j.issn.1001-6600.2016.03.006.
[5]ELLAHYANI A, EL ANSARI M, EL JAAFARI I. Traffic sign detection and recognition based on random forests[J]. Applied Soft Computing, 2015, 46: 805-815. DOI: 10.1016/j.asoc.2015.12.041.
[6]徐岩,王权威,韦镇余.一种融合加权ELM和AdaBoost的交通标志识别算法[J].小型微型计算机系统,2017,38(9):2028-2032.
[7]何广焕,唐诗华,邢鹏威,等. 基于WOA-LSSVM算法的区域GNSS 高程异常拟合方法[J].桂林理工大学学报,2021,41(4):837-842.
[8]柴润泽,叶松,熊伟. 基于形状特征的花椒分选[J].桂林电子科技大学学报,2021,41(5):382-386.DOI: 10.16725/j.cnki.cn45-1351/tn.2021.05.007.
[9]韩习习,魏民,徐西义,等.基于多特征融合的交通标志识别算法[J].计算机工程与应用, 2019,55(18): 195-200. DOI: 10.3778/j.issn.1002-8331.1806-0155.
[10]DA SILVA D L, SEIJAS L M, BASTOS-FILHO C J A. Artificial bee colony optimization for feature selection of traffic sign recognition[J]. International Journal of Swarm Intelligence Research, 2017, 8(2): 50-66. DOI: 10.4018/IJSIR.2017040104.
[11]XU Y, WANG Q W, WEI Z Y, et al. Traffic sign recognition based on weighted ELM and AdaBoost[J]. Electronics Letters, 2016, 52(24): 1988-1990. DOI: 10.1049/el.2016.2299.
[12]TIMBU C, MICLEA V, LEMNARU C. Semantic segmentation-based traffic sign detection and recognition using deep learning techniques[C]// 2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP). Piscataway,NJ:IEEE, 2018: 325-331. DOI: 10.1109/ICCP.2018.8516600.
[13]唐熔钗,伍锡如.基于改进YOLO-V3网络的百香果实时检测[J].广西师范大学学报(自然科学版),2020,38(6):32-39. DOI: 10.16088/j.issn.1001-6600.2020.06.004.
[14]STEPHEN O, SAIN M, MADUH U J, et al. An efficient deep learning approach to pneumonia classification in healthcare[J]. Journal of Healthcare Engineering, 2019,2019:4180949. DOI: 10.1155/2019/4180949.
[15]柯鹏飞,蔡茂国,吴涛.基于改进卷积神经网络与集成学习的人脸识别算法[J].计算机工程,2020,46(2):262-267,273. DOI: 10.19678/j.issn.1000-3428.0053576.
[16]LI L F, WANG Y R. Improved LeNet-5 convolutional neural network traffic sign recognition[J]. International Core Journal of Engineering, 2021, 7(4): 114-121. DOI: 10.6919/ICJE.202104_7(4).0017.
[17]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. DOI: 10.1016/j.neunet.2012.02.016.
[18]ZHANG J M, WANG W, LU C Q, et al. Lightweight deep network for traffic sign classification[J]. Annals of Telecommunications, 2020, 75(7): 369-379. DOI: 10.1007/s12243-019-00731-9.
[19]ZHOU S R, LIANG W L, LI J G, et al. Improved VGG model for road traffic sign recognition[J]. Computers, Materials & Continua, 2018, 57(1): 11-24. DOI: 10.32604/cmc.2018.02617.
[20]HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway,NJ:IEEE, 2016: 770-778. DOI:10.1109/CVPR.2016.90.
[21]邓文轩,杨航,靳婷.基于注意力机制的图像分类降维方法[J].广西师范大学学报(自然科学版),2021,39(2):32-40. DOI: 10.16088/j.issn.1001-6600.2020090704.
[22]GMEZ SERNA C, RUICHEK Y. Classification of traffic signs: the European dataset[J]. IEEE Access, 2018, 6: 78136-78148.DOI: 10.1109/ACCESS.2018.2884826.
[23]XU H, SRIVASTAVA G. Automatic recognition algorithm of traffic signs based on convolution neural network[J]. Multimedia Tools and Applications, 2020, 79(17): 11551-11565.DOI: 10.1007/s11042-019-08239-z.
[24]SAPIJASZKO G, ALOBAIDI T, MIKHAEL W B. Traffic sign recognition based on multilayer perceptron using DWT and DCT[C]// 2019 IEEE 62nd International Midwest Symposium on Circuits and Systems (MWSCAS). Piscataway,NJ:IEEE, 2019: 440-443. DOI: 10.1109/MWSCAS.2019.8884897.
[25]邝先验,朱磊,吴赟,等.基于Adaboost算法和颜色特征的公交车辆视频检测[J].广西师范大学学报(自然科学版),2016,34(1):9-18. DOI: 10.16088/j.issn.1001-6600.2016.01.002.
[26]TONG G F, CHEN H R, LI Y, et al. Traffic sign recognition based on SVM and convolutional neural network[C]// 2017 12th IEEE Conference on Industrial Electronics and Applications (ICIEA). Piscataway,NJ:IEEE, 2017: 2066-2071. DOI: 10.1109/ICIEA.2017.8283178.
[27]张功国,吴建,易亿,等.基于集成卷积神经网络的交通标志识别[J].重庆邮电大学学报(自然科学版),2019,31(4):571-577.
[28]屈治华,王琳,邵毅明,等.一种面向多视角交通标志识别的CapsNet集成算法[J].重庆理工大学学报(自然科学),2020,34(10):160-168.
[29]PALAK,SANGAL A L. Traffic signs classification using convolutional neural networks: a review[C]// 2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC). Piscataway,NJ:IEEE, 2021: 450-455. DOI: 10.1109/ICSCCC51823.2021.9478172.

相关文章 14

[1]	张喜龙, 韩萌, 陈志强, 武红鑫, 李慕航. 面向复杂数据流的集成分类综述[J]. 广西师范大学学报（自然科学版）, 2022, 40(4): 1-21.
[2]	周圣凯, 富丽贞, 宋文爱. 基于深度学习的短文本语义相似度计算模型[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 49-56.
[3]	彭涛, 唐经, 何凯, 胡新荣, 刘军平, 何儒汉. 基于多步态特征融合的情感识别[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 104-111.
[4]	胡强, 刘倩, 周杭霞. 基于改进Stacking策略的钓鱼网站检测研究[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 132-140.
[5]	马铖旭, 曾上游, 赵俊博, 陈红阳. 基于卷积神经网络的逆光图像增强研究[J]. 广西师范大学学报（自然科学版）, 2022, 40(2): 81-90.
[6]	陈文康, 陆声链, 刘冰浩, 李帼, 刘晓宇, 陈明. 基于改进YOLOv4的果园柑橘检测方法研究[J]. 广西师范大学学报（自然科学版）, 2021, 39(5): 134-146.
[7]	杨州, 范意兴, 朱小飞, 郭嘉丰, 王越. 神经信息检索模型建模因素综述[J]. 广西师范大学学报（自然科学版）, 2021, 39(2): 1-12.
[8]	邓文轩, 杨航, 靳婷. 基于注意力机制的图像分类降维方法[J]. 广西师范大学学报（自然科学版）, 2021, 39(2): 32-40.
[9]	白捷, 高海力, 王永众, 杨来邦, 项晓航, 楼雄伟. 基于多路特征融合的Faster R-CNN与迁移学习的学生课堂行为检测[J]. 广西师范大学学报（自然科学版）, 2020, 38(5): 1-11.
[10]	严浩, 许洪波, 沈英汉, 程学旗. 开放式中文事件检测研究[J]. 广西师范大学学报（自然科学版）, 2020, 38(2): 64-71.
[11]	范瑞,蒋品群,曾上游,夏海英,廖志贤,李鹏. 多尺度并行融合的轻量级卷积神经网络设计[J]. 广西师范大学学报（自然科学版）, 2019, 37(3): 50-59.
[12]	武文雅, 陈钰枫, 徐金安, 张玉洁. 基于高层语义注意力机制的中文实体关系抽取[J]. 广西师范大学学报（自然科学版）, 2019, 37(1): 32-41.
[13]	薛洋,曾庆科,夏海英,王文涛. 基于卷积神经网络超分辨率重建的遥感图像融合[J]. 广西师范大学学报（自然科学版）, 2018, 36(2): 33-41.
[14]	孙水发, 李乐鹏, 董方敏, 邹耀斌, 陈鹏. 基于迭代阈值的子块部分重叠双直方图均衡算法[J]. 广西师范大学学报（自然科学版）, 2013, 31(3): 119-126.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed