TY - JOUR
T1 - Semisupervised PCA convolutional network for vehicle type classification
AU - Soon, Foo Chong
AU - Khaw, Hui Ying
AU - Chuah, Joon Huang
AU - Kanesan, Jeevan
N1 - Funding Information:
Manuscript received December 30, 2018; revised May 1, 2019, August 22, 2019, December 16, 2019, and February 17, 2020; accepted May 10, 2020. Date of publication June 5, 2020; date of current version August 13, 2020. This work was supported by the Research Fund Assistance (BKP) Grant from the University of Malaya under Grant BKS101-2017. The review of this article was coordinated by Prof. M. S. Ahmed. (Corresponding author: Joon Huang Chuah.) The authors are with the Department of Electrical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia (e-mail: josephsoonfc@gmail.com; huiyingkhaw@gmail.com; jhchuah@um.edu.my; jievan@um.edu.my). Digital Object Identifier 10.1109/TVT.2020.3000306
Publisher Copyright:
© 1967-2012 IEEE.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/8
Y1 - 2020/8
N2 - In the vehicle type classification area, the necessity to improve classification performance across traffic surveillance cameras has garnered attention in research especially on high level feature extraction and classification. The backpropagation (BP) training approach of traditional deep Convolutional Neural Network (CNN) approach is time-consuming without using a Graphics Processing Unit (GPU). In this paper, we propose a semisupervised strategy for the end-To-end Principal Component Analysis Convolutional Network (PCN) in the area of vehicle type classification. Even without using a GPU, the proposed model eliminates the time-consuming training procedure of convolutional filter bank. In particular, the convolutional filters of the network are generated using unsupervised learning by Principal Component Analysis (PCA) which has tremendously reduced training cost and also reinforced the robustness of extracted features against various distortions. In order to further improve the training procedure while still preserving the discriminative characteristic of the system, only the fully-connected layer is fine-Tuned in the supervised classification stage. The PCN is tested using a public BIT-Vehicle dataset which comprises 9850 surveillance-nature vehicle frontal-view images. The PCN can be easily implemented and readily compatible with many effective classifiers. Two classifiers, namely softmax classifier and Support Vector Machine (SVM) are employed in this network and their classification performances are then compared. Both classifiers take less than 100 seconds in the training process and are able to produce an average accuracy of above 88.35%, even under various inferior imaging conditions.
AB - In the vehicle type classification area, the necessity to improve classification performance across traffic surveillance cameras has garnered attention in research especially on high level feature extraction and classification. The backpropagation (BP) training approach of traditional deep Convolutional Neural Network (CNN) approach is time-consuming without using a Graphics Processing Unit (GPU). In this paper, we propose a semisupervised strategy for the end-To-end Principal Component Analysis Convolutional Network (PCN) in the area of vehicle type classification. Even without using a GPU, the proposed model eliminates the time-consuming training procedure of convolutional filter bank. In particular, the convolutional filters of the network are generated using unsupervised learning by Principal Component Analysis (PCA) which has tremendously reduced training cost and also reinforced the robustness of extracted features against various distortions. In order to further improve the training procedure while still preserving the discriminative characteristic of the system, only the fully-connected layer is fine-Tuned in the supervised classification stage. The PCN is tested using a public BIT-Vehicle dataset which comprises 9850 surveillance-nature vehicle frontal-view images. The PCN can be easily implemented and readily compatible with many effective classifiers. Two classifiers, namely softmax classifier and Support Vector Machine (SVM) are employed in this network and their classification performances are then compared. Both classifiers take less than 100 seconds in the training process and are able to produce an average accuracy of above 88.35%, even under various inferior imaging conditions.
KW - Convolutional neural network
KW - principal component analysis
KW - softmax
KW - support vector machine
KW - vehicle type classification
UR - http://www.scopus.com/inward/record.url?scp=85090164031&partnerID=8YFLogxK
U2 - 10.1109/TVT.2020.3000306
DO - 10.1109/TVT.2020.3000306
M3 - Article
AN - SCOPUS:85090164031
VL - 69
SP - 8267
EP - 8277
JO - IEEE Transactions on Vehicular Technology
JF - IEEE Transactions on Vehicular Technology
SN - 0018-9545
IS - 8
M1 - 9109741
ER -