基于注意力引导数据增强的车型识别

doi:10.19678/j.issn.1000-3428.0062096

计算机工程 ›› 2022, Vol. 48 ›› Issue (7): 300-306. doi: 10.19678/j.issn.1000-3428.0062096

基于注意力引导数据增强的车型识别

孙伟^1,2, 常鹏帅¹, 戴亮¹, 张小瑞^2,3,4, 陈旋⁴, 代广昭¹

1. 南京信息工程大学自动化学院, 南京 210044;
2. 南京信息工程大学江苏省大气环境与装备技术协同创新中心, 南京 210044;
3. 南京信息工程大学数字取证教育部工程研究中心, 南京 210044;
4. 南京信息工程大学计算机与软件学院, 南京 210044

收稿日期:2021-07-15 修回日期:2021-08-18 出版日期:2022-07-15 发布日期:2021-08-24
作者简介:孙伟(1980—),男,副教授、博士,主研方向为模式识别、图像处理、车辆视频检测与识别;常鹏帅、戴亮,硕士研究生;张小瑞,教授、博士;陈旋、代广昭,硕士研究生。
基金资助:
国家自然科学基金（61304205）；江苏省自然科学基金（BK20191401，BK20201136）；江苏省研究生科研与实践创新计划项目（SJCX21_0363）；大学生创新创业训练项目（202010300290，202010300211，202010300116E）。

Vehicle Type Recognition Based on Attention Guided Data Augmentation

SUN Wei^1,2, CHANG Pengshuai¹, DAI Liang¹, ZHANG Xiaorui^2,3,4, CHEN Xuan⁴, DAI Guangzhao¹

1. School of Automation, Nanjing University of Information Science and Technology, Nanjing 210044, China;
2. Jiangsu Province Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing University of Information Science and Technology, Nanjing 210044, China;
3. Engineering Research Center of Digital Forensics of Ministry of Education, Nanjing University of Information Science and Technology, Nanjing 210044, China;
4. School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing 210044, China

Received:2021-07-15 Revised:2021-08-18 Online:2022-07-15 Published:2021-08-24

摘要/Abstract

摘要： 车型识别在智能交通系统中发挥着重要作用。受车辆数据不足、车辆类间差异小等因素的影响，传统车型识别方法未充分利用车辆鉴别性区域的特征，导致识别准确率降低。提出一种基于注意力模块引导数据增强的车型识别方法。将ResNet-50作为骨干网络提取车辆特征，同时在网络的每个残差块后均嵌入坐标注意力模块，编码成一对方向感知和位置敏感的注意力图，以增强车辆鉴别性区域的特征表达。在此基础上，利用双线性注意力汇集操作生成增强特征图，通过对增强特征图进行注意力裁剪和注意力擦除，获取具有强鉴别性的增强数据。在Stanford Cars车辆数据集上的实验结果验证了该方法的有效性，结果表明，该方法的车型识别准确率达到94.86%，与RA-CNN、MA-CNN、WS-DAN+Inception-v3等方法相比，能够有效提高车型识别准确率和数据增强效率。

关键词: 车型识别, 坐标注意力, 数据增强, 注意力裁剪, 注意力擦除

Abstract: Vehicle type recognition plays an important role in intelligent transportation systems.Owing to the lack of vehicle data and small differences between vehicle classes, traditional vehicle type recognition do not make full use of the features of the vehicle discriminant area, resulting in a reduction in recognition accuracy.This study proposes a vehicle type recognition method based on attention guided data augmentation.In this method, ResNet-50 is used as the backbone network to extract vehicle features.Simultaneously, a Coordinate Attention(CA) module is embedded behind each residual block of the network to encode a pair of direction-aware and position-sensitive attention diagrams to enhance the feature representation of the vehicle discriminant area.On this basis, the bilinear attention-gathering operation is used to effectively obtain the enhanced feature image.Through attention cropping and erasure of the enhanced feature map, enhanced data with strong discrimination are obtained.The results on the Stanford Cars vehicle dataset verify the effectiveness of this method.The results showed that the accuracy of vehicle type recognition of this method reaches 94.86%.Compared with RA-CNN, MA-CNN, WS-DAN+Inception-v3, and other methods, it can effectively improve the accuracy of vehicle type recognition and efficiency of data augmentation.

Key words: vehicle type recognition, Coordinate Attention(CA), data augmentation, attention cropping, attention erasure

中图分类号:

TP391

孙伟, 常鹏帅, 戴亮, 张小瑞, 陈旋, 代广昭. 基于注意力引导数据增强的车型识别[J]. 计算机工程, 2022, 48(7): 300-306.

SUN Wei, CHANG Pengshuai, DAI Liang, ZHANG Xiaorui, CHEN Xuan, DAI Guangzhao. Vehicle Type Recognition Based on Attention Guided Data Augmentation[J]. Computer Engineering, 2022, 48(7): 300-306.

http://www.ecice06.com/CN/Y2022/V48/I7/300

图/表 11

20221029180031

20221029180034

20221029180050

20221029180053

20221029180058

20221029180101

20221029180105

20221029180109

20221029180112

20221029180115

20221029180119

参考文献

[1] SRIDEVI T, SWAPNA P, HARINATH K.Vehicle identification based on the model[C]//Proceedings of the 7th International Advance Computing Conference.Washington D.C., USA:IEEE Press, 2017:566-571.
[2] YANG L J, LUO P, CHANGE L C, et al.A large- scale car dataset for fine-grained categorization and verification[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2015:3973-3981.
[3] 罗建豪, 吴建鑫.基于深度卷积特征的细粒度图像分类研究综述[J].自动化学报, 2017, 43(8):1306-1318. LUO J H, WU J X.A survey on fine-grained image categorization using deep convolutional features[J].Acta Automatica Sinica, 2017, 43(8):1306-1318.(in Chinese)
[4] 张小瑞, 陈旋, 孙伟, 等.基于深度学习的车辆再识别研究进展[J].计算机工程, 2020, 46(11):1-11. ZHANG X R, CHEN X, SUN W, et al.Progress of vehicle re-identification research based on deep learning[J]. Computer Engineering, 2020, 46(11):1-11.(in Chinese)
[5] 郑远攀, 李广阳, 李晔.深度学习在图像识别中的应用研究综述[J].计算机工程与应用, 2019, 55(12):20-36. ZHENG Y P, LI G Y, LI Y.Survey of application of deep learning in image recognition[J].Computer Engineering and Applications, 2019, 55(12):20-36.(in Chinese)
[6] HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:770-778.
[7] HOU Q B, ZHANG L, CHENG M M, et al.Strip pooling:rethinking spatial pooling for scene parsing[EB/OL].[2021-06-10].https://arxiv.org/abs/2003.13328.
[8] 叶晴昊, 涂岱键, 毕奇, 等.基于多视图架构深度神经网络的图像威胁识别[J].计算机工程, 2020, 46(11):261-266. YE Q H, TU D J, BI Q, et al.Image threat recognition based on multiple view architecture deep neural network[J]. Computer Engineering, 2020, 46(11):261-266.(in Chinese)
[9] 张儒鹏, 于亚新, 张康, 等.基于OI-LSTM神经网络结构的人类动作识别模型研究[J].计算机科学与探索, 2018, 12(12):1926-1939. ZHANG R P, YU Y X, ZHANG K, et al.Research on human action recognition model based on OI-LSTM neural network structure[J].Journal of Frontiers of Computer Science and Technology, 2018, 12(12):1926-1939.(in Chinese)
[10] JIE H, LI S, SAMUEL A, et al.Squeeze-and-excitation networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:7132-7141.
[11] WOO S, PARK J, LEE J Y, et al.CBAM:convolutional block attention module[EB/OL].[2021-06-10].https://arxiv.org/abs/1807.06521.
[12] HOU Q B, ZHOU D Q, FENG J S.Coordinate attention for efficient mobile network design[EB/OL].[2021-06-10].https://arxiv.org/abs/2103.02907.
[13] DEVRIES T, TAYLOR G W.Improved regularization of convolutional neural networks with cutout[EB/OL].[2021-06-10].https://arxiv.org/abs/1708.04552.
[14] ZHONG Z, ZHENG L, KANG G L, et al.Random erasing data augmentation[EB/OL].[2021-06-10].https://arxiv.org/abs/1708.04896.
[15] CUBUK E D, ZOPH B, MANE D, et al.Autoaugment:learning augmentation policies from data[EB/OL].[2021-06-10].https://arxiv.org/abs/1805.09501.
[16] HU T, QI H G, HUANG Q M, et al.See better before looking closer:weakly supervised data augmentation network for fine-grained visual classification[EB/OL].[2021-06-10].https://arxiv.org/abs/1901.09891.
[17] WEN Y, ZHANG K, LI Z, et al.A discriminative feature learning approach for deep face recognition[C]//Proceedings of European Conference on Computer Vision.Berlin, Germany:Springer, 2016:499-515.
[18] KRAUSE J, STARK M, DENG J, et al.3D object representations for fine-grained categorization[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2013:554-561.
[19] SELVARAJU R R, COGSWELL M, DAS A, et al.Grad-CAM:visual explanations from deep networks via gradient-based localization[J].International Journal of Computer Vision, 2020, 128(2):336-359.
[20] SIMONYAN K, ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[EB/OL].[2021-06-10].https://arxiv.org/pdf/1409.1556.pdf.
[21] SZEGEDY C, VANHOUCKE V, IOFFE S, et al.Rethinking the inception architecture for computer vision[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:1-10.
[22] FU J, ZHENG H, TAO M.Look closer to see better:recurrent attention convolutional neural network for fine-grained image recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2017:4476-4484.
[23] ZHENG H, FU J, TAO M, et al.Learning multi-attention convolutional neural network for fine-grained image recognition[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:5219-5227.

选择文件类型/文献管理软件名称

选择包含的内容

基于注意力引导数据增强的车型识别

Vehicle Type Recognition Based on Attention Guided Data Augmentation

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	江雨燕, 陶承凤, 李平. 数据增强和自适应自步学习的深度子空间聚类算法[J]. 计算机工程, 2023, 49(8): 96-103, 110.
[2]	席荣康, 蔡满春, 芦天亮. 基于数据增强与流数据处理的Tor流量分析模型[J]. 计算机工程, 2023, 49(3): 177-184.
[3]	王禹博, 陈利锋, 许卫霞. 结合多解码器与两阶段通道选择的异常检测方法[J]. 计算机工程, 2023, 49(3): 37-48.
[4]	郑伟鹏, 罗晓曙, 蒙志明. 基于改进轻量级秩扩展网络的人脸表情识别方法[J]. 计算机工程, 2022, 48(9): 189-196.
[5]	曹瑞阳, 郭佑民, 牛满宇. 基于最大最小距离的多中心数据综合增强方法[J]. 计算机工程, 2022, 48(6): 174-181.
[6]	佘朝阳, 严馨, 徐广义, 陈玮, 邓忠莹. 融合数据增强与半监督学习的药物不良反应检测[J]. 计算机工程, 2022, 48(6): 314-320.
[7]	鲁统伟, 徐子昕, 闵锋. 基于生成对抗网络的知识蒸馏数据增强[J]. 计算机工程, 2022, 48(4): 70-80.
[8]	乔栋, 陈章进, 邓良, 屠程力. 基于改进语音处理的卷积神经网络中文语音情感识别方法[J]. 计算机工程, 2022, 48(2): 281-290.
[9]	谭润, 叶武剑, 刘怡俊. 结合双语义数据增强与目标定位的细粒度图像分类[J]. 计算机工程, 2022, 48(2): 237-242,249.
[10]	刘圣杰, 何宁, 于海港, 王程, 韩文静. 引入坐标注意力和自注意力的人体关键点检测研究[J]. 计算机工程, 2022, 48(12): 86-94.
[11]	范馨月, 鲍泓, 潘卫国. 基于类别不平衡数据集的图像实例分割方法[J]. 计算机工程, 2022, 48(12): 224-231.
[12]	陈良臣, 傅德印. 面向小样本数据的机器学习方法研究综述[J]. 计算机工程, 2022, 48(11): 1-13.
[13]	代广昭, 孙伟, 徐凡, 张小瑞, 陈旋, 常鹏帅, 汤毅, 胡亚华. 用于车辆重识别的视角感知局部注意力网络[J]. 计算机工程, 2022, 48(10): 288-297,305.
[14]	程广涛, 巩家昌, 赵洪伟. 基于膨胀卷积和稠密连接的烟雾识别方法[J]. 计算机工程, 2020, 46(4): 253-259.
[15]	蒋芸,谭宁,张海,彭婷婷. 基于条件生成对抗网络的咬翼片图像分割[J]. 计算机工程, 2019, 45(4): 223-227.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于注意力引导数据增强的车型识别

Vehicle Type Recognition Based on Attention Guided Data Augmentation

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献

相关文章 15

编辑推荐

Metrics

本文评价