首页 » 文章 » 文章详细信息
Advances in Multimedia Volume 2018 ,2018-10-02
Pretraining Convolutional Neural Networks for Image-Based Vehicle Classification
Research Article
Yunfei Han 1 , 2 , 3 Tonghai Jiang 1 , 2 , 3 Yupeng Ma 1 , 2 , 3 Chunxiang Xu 1 , 2 , 3
Show affiliations
DOI:10.1155/2018/3138278
Received 2018-05-18, accepted for publication 2018-09-13, Published 2018-09-13
PDF
摘要

Vehicle detection and classification are very important for analysis of vehicle behavior in intelligent transportation system, urban computing, etc. In this paper, an approach based on convolutional neural networks (CNNs) has been applied for vehicle classification. In order to achieve a more accurate classification, we removed the unrelated background as much as possible based on a trained object detection model. In addition, an unsupervised pretraining approach has been introduced to better initialize CNNs parameters to enhance the classification performance. Through the data enhancement on manual labeled images, we got 2000 labeled images in each category of motorcycle, transporter, passenger, and others, with 1400 samples for training and 600 samples for testing. Then, we got 17395 unlabeled images for layer-wise unsupervised pretraining convolutional layers. A remarkable accuracy of 93.50% is obtained, demonstrating the high classification potential of our approach.

授权许可

Copyright © 2018 Yunfei Han et al. 2018
This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

通讯作者

Tonghai Jiang.The Xinjiang Technical Institute of Physics & Chemistry, Urumqi 830011, China;Xinjiang Laboratory of Minority Speech and Language Information Processing, Urumqi 830011, China;University of Chinese Academy of Sciences, Beijing 100049, China, ucas.ac.cn.jth@ms.xjb.ac.cn

推荐引用方式

Yunfei Han,Tonghai Jiang,Yupeng Ma,Chunxiang Xu. Pretraining Convolutional Neural Networks for Image-Based Vehicle Classification. Advances in Multimedia ,Vol.2018(2018)

您觉得这篇文章对您有帮助吗?
分享和收藏
0

是否收藏?

参考文献
[1] N. Dalal, B. Triggs. Histograms of oriented gradients for human detection. .1:886-893. DOI: 10.1145/2808492.2808495.
[2] Y. Jo, I. Jung. (2014). Analysis of vehicle detection with wsn-based ultrasonic sensors. Sensors.14(8):14050-14069. DOI: 10.1145/2808492.2808495.
[3] J. Dai, Y. Li, K. He, J. Sun. et al.R-FCN: Object Detection via Region-based Fully Convolutional Networks. . DOI: 10.1145/2808492.2808495.
[4] M. D. Zeiler, D. Krishnan, G. W. Taylor, R. Fergus. et al.Deconvolutional networks. :2528-2535. DOI: 10.1145/2808492.2808495.
[5] J. Redmon, A. Farhadi. YOLO9000: Better, faster, stronger. :6517-6525. DOI: 10.1145/2808492.2808495.
[6] Y. Iwasaki, M. Misumi, T. Nakamiya. (2015). Robust vehicle detection under various environments to realize road traffic flow surveillance using an infrared thermal camera. The Scientific World Journal.2015. DOI: 10.1145/2808492.2808495.
[7] R. A. Hadi, G. Sulong, L. E. George. (2014). Vehicle Detection and Tracking Techniques: A Concise Review. Signal & Image Processing: An International Journal.5(1):1-12. DOI: 10.1145/2808492.2808495.
[8] R. Girshick, J. Donahue, T. Darrell, J. Malik. et al.Rich feature hierarchies for accurate object detection and semantic segmentation. :580-587. DOI: 10.1145/2808492.2808495.
[9] J. F. Forren, D. Jaarsma. (1999). Traffic monitoring by tire noise. Computer Standards & Interfaces.20:466-467. DOI: 10.1145/2808492.2808495.
[10] J. Y. Ng, Y. H. Tay. Image-based Vehicle Classification System. . DOI: 10.1145/2808492.2808495.
[11] G. E. Hinton, R. S. Zemel. (1994). Autoencoders, minimum description length and helmholtz free energy. Neural Information Processing Systems:3-10. DOI: 10.1145/2808492.2808495.
[12] A. Duzdar, G. Kompa. Applications using a low-cost baseband pulsed microwave radar sensor. .1:239-243. DOI: 10.1145/2808492.2808495.
[13] W. Liu, D. Anguelov, D. Erhan, C. Szegedy. et al.SSD: single shot multibox detector. .9905:21-37. DOI: 10.1145/2808492.2808495.
[14] F. M. Kazemi, S. Samadi, H. R. Poorreza, M.-R. Akbarzadeh-T. et al.Vehicle recognition based on fourier, wavelet and curvelet transforms - A comparative study. :939-940. DOI: 10.1145/2808492.2808495.
[15] H.-T. Kim, B. Song. Vehicle recognition based on radar and vision sensor fusion for automatic emergency braking. :1342-1346. DOI: 10.1145/2808492.2808495.
[16] D. He, C. Lang, S. Feng, X. Du. et al.Vehicle detection and classification based on convolutional neural network. . DOI: 10.1145/2808492.2808495.
[17] X. Wen, L. Shao, Y. Xue, W. Fang. et al.(2015). A rapid learning algorithm for vehicle classification. Information Sciences.295:395-406. DOI: 10.1145/2808492.2808495.
[18] P. Sermanet, D. Eigen, X. Zhang, M. Mathieu. et al.OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks. . DOI: 10.1145/2808492.2808495.
[19] Z. Chen, T. Ellis, S. A. Velastin. Vehicle detection, tracking and classification in urban traffic. :951-956. DOI: 10.1145/2808492.2808495.
[20] B. Li, T. Zhang, T. Xia. Vehicle detection from 3D lidar using fully convolutional network. . DOI: 10.1145/2808492.2808495.
[21] Y. Lecun, B. Boser, J. S. Denker, D. Henderson. et al.(1989). Backpropagation applied to handwritten zip code recognition. Neural Computation.1(4):541-551. DOI: 10.1145/2808492.2808495.
[22] A. Krizhevsky, I. Sutskever, G. E. Hinton. (2012). ImageNet classification with deep convolutional neural networks. Neural Information Processing Systems:1097-1105. DOI: 10.1145/2808492.2808495.
[23] J. Arróspide, L. Salgado. (2014). A study of feature combination for vehicle detection based on image processing. The Scientific World Journal.2014-13. DOI: 10.1145/2808492.2808495.
[24] Y. Lecun, L. Bottou, Y. Bengio, P. Haffner. et al.(1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE.86(11):2278-2323. DOI: 10.1145/2808492.2808495.
[25] P. Piyush, R. Rajan, L. Mary, B. I. Koshy. et al.Vehicle detection and classification using audio-visual cues. :726-730. DOI: 10.1145/2808492.2808495.
[26] J. Masci, U. Meier, D. Cireşan, J. Schmidhuber. et al.Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction. .6791:52-59. DOI: 10.1145/2808492.2808495.
[27] K. Wang, R. Wang, Y. Feng, H. Zhang. et al.Vehicle recognition in acoustic sensor networks via sparse representation. :1-4. DOI: 10.1145/2808492.2808495.
[28] M. Yi, F. Yang, E. Blashch, C. Sheaff. et al.(2016). Vehicle Classification in WAMI Imagery using Deep Network. Proceedings of the SPIE 9838: Sensors and Systems for Space Applicatioins IX. DOI: 10.1145/2808492.2808495.
[29] R. Girshick. Fast R-CNN. :1440-1448. DOI: 10.1145/2808492.2808495.
[30] J. George, L. Mary, K. S. Riyas. Vehicle detection and classification from acoustic signal using ANN and KNN. :436-439. DOI: 10.1145/2808492.2808495.
[31] Y. Bengio, P. Lamblin, D. Popovici, H. Larochelle. et al.(2007). Greedy layer-wise training of deep networks. Neural Information Processing Systems:153-160. DOI: 10.1145/2808492.2808495.
[32] S. Ren, K. He, R. Girshick, J. Sun. et al.(2017). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence.39(6):1137-1149. DOI: 10.1145/2808492.2808495.
[33] V. Kastrinaki, M. Zervakis, K. Kalaitzakis. (2003). A survey of video processing techniques for traffic applications. Image and Vision Computing.21(4):359-381. DOI: 10.1145/2808492.2808495.
[34] E. Rublee, V. Rabaud, K. Konolige, G. Bradski. et al.ORB: an efficient alternative to SIFT or SURF. :2564-2571. DOI: 10.1145/2808492.2808495.
[35] E. Tola, V. Lepetit, P. Fua. (2010). DAISY: an efficient dense descriptor applied to wide-baseline stereo. IEEE Transactions on Pattern Analysis and Machine Intelligence.32(5):815-830. DOI: 10.1145/2808492.2808495.
[36] G. E. Hinton, S. Osindero, Y. Teh. (2006). A fast learning algorithm for deep belief nets. Neural Computation.18(7):1527-1554. DOI: 10.1145/2808492.2808495.
[37] J. Redmon, S. Divvala, R. Girshick, A. Farhadi. et al.You only look once: Unified, real-time object detection. :779-788. DOI: 10.1145/2808492.2808495.
[38] D. G. Lowe. Object recognition from local scale-invariant features. .2:1150-1157. DOI: 10.1145/2808492.2808495.
[39] B. Li, T. Zhang, T. Xia. Vehicle Detection from 3D Lidar Using Fully Convolutional Network. . DOI: 10.1145/2808492.2808495.
[40] D. Erhan, Y. Bengio, A. Courville, P.-A. Manzagol. et al.(2010). Why does unsupervised pre-training help deep learning?. Journal of Machine Learning Research.11:625-660. DOI: 10.1145/2808492.2808495.
[41] J. Lan, Y. Xiang, L. Wang, Y. Shi. et al.(2011). Vehicle detection and classification by measuring and processing magnetic signal. Measurement.44(1):174-180. DOI: 10.1145/2808492.2808495.
[42] J. George, A. Cyril, B. I. Koshy, L. Mary. et al.(2013). Exploring Sound Signature for Vehicle Detection and Classification Using ANN. International Journal on Soft Computing.4(2):29-36. DOI: 10.1145/2808492.2808495.
[43] H. Wang, Y. Cai, L. Chen. (2014). A vehicle detection algorithm based on deep belief network. The Scientific World Journal.2014-7. DOI: 10.1145/2808492.2808495.
[44] A. Tourani, A. Shahbahrami. Vehicle counting method based on digital image processing algorithms. :1-6. DOI: 10.1145/2808492.2808495.
[45] P. Mishra, B. Banerjee. (2013). Multiple Kernel based KNN Classifiers for Vehicle Classification. International Journal of Computer Applications.71(6):1-7. DOI: 10.1145/2808492.2808495.
文献评价指标
浏览 307次
下载全文 37次
评分次数 1次
用户评分 5.0分
分享 0次