[1]费东炜,孙 涵.基于深度哈希网络的车型识别方法[J].计算机技术与发展,2020,30(01):7-12.[doi:10. 3969 / j. issn. 1673-629X. 2020. 01. 002]
　FEI Dong-wei,SUN Han.A Vehicle Recognition Algorithm Based on Deep Hashing Network[J].Computer Technology and Development,2020,30(01):7-12.[doi:10. 3969 / j. issn. 1673-629X. 2020. 01. 002]
点击复制

基于深度哈希网络的车型识别方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年01期
页码:
7-12
栏目:
智能、算法、系统工程
出版日期:
2020-01-10

文章信息/Info

Title:
A Vehicle Recognition Algorithm Based on Deep Hashing Network
文章编号:
1673-629X(2020)01-0007-06
作者:
费东炜; 孙 涵
南京航空航天大学 计算机科学与技术学院 模式分析与机器智能工业和信息化部重点实验室,江苏 南京 211106
Author(s):
FEI Dong-wei; SUN Han
MIIT Key Laboratory of Pattern Analysis and Machine Intelligence,School of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China
关键词:
车型识别; 卷积神经网络; 数据增广; 全局平均池化; 深度哈希网络
Keywords:
vehicle recognition; convolution neural network; data augmentation; global average pooling; deep hashing network
分类号:
TP39
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 01. 002
摘要:
针对车型识别任务的特点,设计了一种基于深度哈希网络的车型识别方法,实现了在类间差异不明显、样本量较少的情况下进行车型检索和分类。 对数据增广方法进行研究,针对车型数据集的特点,提出了适用于车型识别的数据增广方法,有效提升了小样本车型识别的准确率。 深度哈希网络采用改进的 HashNet 网络来快速学习车辆的二值特征表达,针对深度哈希网络使用全连接层导致参数过多的问题,提出了 HashNet-GAP 网络,以全局平均池化层替换了 HashNet 中的部分全连接层。 相对于 HashNet 网络,大幅度减少了参数数量,提升了前向计算速度和网络性能。 实验结果表明,该车型识别方法能够对类间差距很小的不同车型进行有效识别,在小样本数据集上取得 80.0%的 Top1 准确率,并且能够显著降低模型的存储消耗和内存消耗。
Abstract:
According to the characteristics of vehicle recognition tasks,a vehicle recognition method based on deep hashing network is designed,which realizes the retrieval and classification of vehicles when the differences between the classes are not obvious and the size of training set is small.The data augmentation method is studied.According to the characteristics? of the vehicle data set, a data augmentation method suitable for vehicle recognition is proposed,which effectively improves the accuracy of small sample recognition. The deep hashing network uses the improved HashNet to quickly learn the binary feature descriptors of the vehicle. The HashNet-GAPnetwork is proposed to reduce the number of parameters caused by full connected layers. The full connected layer is partially replaced by the global average pooling layer. Compared with the HashNet,the number of parameters is greatly reduced,and the forward calculation speed and network performance are improved. The experiment shows that the vehicle recognition method can effectively recognize different vehicle models with small gaps between classes,achieve 80.0% Top1 accuracy on a small dataset,and can significantly reduce storage consumption and memory consumption of the network model.

相似文献/References:

[1]张明恒 王华莹 郭烈.基于改进K—Means算法的车辆识别方法[J].计算机技术与发展,2012,(05):53.
　ZHANG Ming-heng,WANG Hua-ying,GUO Lie.Method of Vehicle Detection Based on Improved K-Means Algorithm[J].Computer Technology and Development,2012,(01):53.
[2]崔凤焦.表情识别算法研究进展与性能比较[J].计算机技术与发展,2018,28(02):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
　CUI Feng-jiao.Ｒesearch and Performance Comparison of Facial Expression Ｒecognition Algorithm[J].Computer Technology and Development,2018,28(01):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
[3]张丹丹,李雷. 基于PCANet-RF的人脸检测系统[J].计算机技术与发展,2016,26(02):31.
　ZHANG Dan-dan,LI Lei. Face Detection System Based on PCANet-RF[J].Computer Technology and Development,2016,26(01):31.
[4]陈强锐,谢世朋.基于深度学习的肺部肿瘤检测方法[J].计算机技术与发展,2018,28(04):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
　CHEN Qiang-rui,XIE Shi-peng.Lung Cancer Detection Method Based on Deep Learning[J].Computer Technology and Development,2018,28(01):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
[5]刘敦强,沈峘,夏瀚笙,等.一种基于深度残差网络的车型识别方法[J].计算机技术与发展,2018,28(05):42.[doi:10.3969/j.issn.1673－629X.2018.05.010]
　LIU Dun-qiang,SHEN Huan,XIA Han-sheng,et al.A Vehicle Model Ｒecognition Algorithm Based on Deep Residual Network[J].Computer Technology and Development,2018,28(01):42.[doi:10.3969/j.issn.1673－629X.2018.05.010]
[6]郭子琰,舒心,刘常燕,等.基于ReLU 函数的卷积神经网络的花卉识别算法[J].计算机技术与发展,2018,28(05):154.[doi:10．3969/j．issn．1673－629X．2018．05．035]
　GUO Ziyan,SHU Xin,LIU Changyan,et al.A Recognition Algorithm of Flower Based on Convolution Neural Network with ReLU Function[J].Computer Technology and Development,2018,28(01):154.[doi:10．3969/j．issn．1673－629X．2018．05．035]
[7]缪宇杰,吴智钧,宫 婧.基于3D 卷积的视频错帧筛选方法[J].计算机技术与发展,2018,28(05):179.[doi:10.3969/ j. issn.1673-629X.2018.05.040]
　MIAO Yu-jie,WU Zhi-jun,GONG Jing.A Wrong Temporal-order Frames Identification Method Based on 3D Convolution[J].Computer Technology and Development,2018,28(01):179.[doi:10.3969/ j. issn.1673-629X.2018.05.040]
[8]李相桥,李晨,田丽华,等.卷积神经网络并行训练的优化研究[J].计算机技术与发展,2018,28(08):12.[doi:10.3969/ j. issn.1673-629X.2018.08.003]
　LI Xiang-qiao,LI Chen,TIAN Li-hua,et al.Research on Optimization of Parallel Training for Convolution Neural Network[J].Computer Technology and Development,2018,28(01):12.[doi:10.3969/ j. issn.1673-629X.2018.08.003]
[9]邓宗平,赵启军,陈虎. 基于深度学习的人脸姿态分类方法[J].计算机技术与发展,2016,26(07):11.
　DEND Zong-ping,ZHAO Qi-jun,CHEN Hu. Face Pose Classification Method Based on Deep Learning[J].Computer Technology and Development,2016,26(01):11.
[10]河海大学 计算机与信息学院,江苏 南京 0098.卷积网络的无监督特征提取对人脸识别的研究[J].计算机技术与发展,2018,28(06):17.[doi:10.3969/ j. issn.1673-629X.2018.06.004]
　DU Bai-sheng.Research on Unsupervised Feature Extraction Based on Convolutional Neural Network for Face Recognition[J].Computer Technology and Development,2018,28(01):17.[doi:10.3969/ j. issn.1673-629X.2018.06.004]
[11]吴玉枝,吴志红,熊运余.基于卷积神经网络的小样本车辆检测与识别[J].计算机技术与发展,2018,28(06):1.[doi:10.3969/ j. issn.1673-629X.2018.06.001]
　WU Yu-zhi,WU Zhi-hong,XIONG Yun-yu.Vehicle Detection and Recognition of a Few Samples Based on Convolutional Neural Network[J].Computer Technology and Development,2018,28(01):1.[doi:10.3969/ j. issn.1673-629X.2018.06.001]

更新日期/Last Update: 2020-01-10