Abstract
Highway monitoring has been an important yet challenging aspect to address with machine learning methods. The method proposed in this paper addresses essential aspects of highway monitoring vehicle detection systems. A machine learning model is proposed to detect cars from dashboard cameras from the datasets. Images used to train the model are collected and grouped from KITTI vision and GTI datasets. The Region-Based Convolutional Neural Network (RCNN) method fails in providing robust information and generates bad candidate region proposals. On the other hand, Faster-RCNN, able to cover the flaws of RCNN, cannot provide good accuracy. Histogram of Oriented Gradients (HOG) features extractor is used over Color Histogram and Spatial Binning as they lack abundant features resulting in a lack of vital information. The method uses vectors from the vehicle and non-vehicles images to improve the classification. Optimized Support Vector Machine (SVM) ‘rbf’ kernel classifier, i.e., Reduced Support Vector Machine (RSVM), excludes ambiguity during classification. The classification was performed using multiple tuned Support Vector Classifiers (SVC), random forest and naive Bayes classifiers. Evaluating with other methods based on accuracy score, average precision and F1-score, RSVM presented the highest performance.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Laopracha, N., Sunat, K., Chiewchanwattana, S.: A novel feature selection in vehicle detection through the selection of dominant patterns of histograms of oriented gradients. IEEE Access 7, 20894–20919 (2019)
Xing, W., Deng, N., Xin, B., Liu, Y., Chen, Y., Zhang, Z.: Identification of extremely similar animal fibers based on matched filter and HOG-SVM. IEEE Access 7, 98603–98617 (2019)
Sun, Z., Bebis, G., Miller, R.: On-road vehicle detection using evolutionary gabor filter optimization. IEEE Trans. Intell. Trans. Syst. 6(2), 125–137 (2005)
Kim, J., Baek, J., Kim, E.: A novel on-road vehicle detection method using HOG. IEEE Trans. Intell. Trans. Syst. 16(6), 3414–3429 (2015)
Liu, C., Hong, F., Liu, R.R., Lu, C.H., Wei, J.: A traffic surveillance multi-scale vehicle detection object method base on encoder-decoder. IEEE Access 8, 47664–47674 (2020)
Xiang, Z., Tan, H., Ye, W.: The excellent properties of a dense grid-based HOG feature on face recognition compared to gabor and LBP. IEEE Access 6, 29306–29319 (2018)
Wang, X.: Data-driven based tiny-YOLOv3 method for front vehicle detection inducing SPP-Net. IEEE Access 8, 110227–110236 (2020)
Zhang, F., Yang, F., Li, C., Yuan, G.: CMNet: a connect-and-merge convolutional neural network for fast vehicle detection in urban traffic surveillance. IEEE Access 7, 72660–72671 (2019)
Wang, Z., Huang, J., Xiong, N.N., Zhou, X., Lin, X., Ward, T.L.: A robust vehicle detection scheme for intelligent traffic surveillance systems in smart cities. IEEE Access 8, 139299–139312 (2020)
Shao, X., Wei, C., Shen, Y., Wang, Z.: Feature enhancement based on CycleGAN for nighttime vehicle detection. IEEE Access 9, 849–859 (2020)
Awasi, M., et al.: Real-Time surveillance through face recognition using HOG And feedforward neural networks. IEEE Access 7, 121236–121244 (2019)
Ghaffari, S., Soleimani, P., Li, K.F., Capson, D.W.: Analysis and comparison of fpga-based histogram of oriented gradients implementations. IEEE Access 8, 79920–79934 (2020)
Hameed, M.A., Hassaballah, M., Ali, S.A., Ismail, A.: An adaptive image steganography method based on histogram of oriented gradient and PVD-LSB techniques. IEEE Access 7, 185189–185204 (2019)
Luo, A., An, F., Zhang, X., Mattausch, H.J.: A Hardware-efficient recognition accelerator using haar-like feature and SVM classifier. IEEE Access 7, 14472–14487 (2019)
Zhang, Y., et al.: Detecting object open angle and direction using machine learning. IEEE Access 8, 12300–12306 (2020)
Wu, G., Chen, W., Cheng, H., Zuo, W., Zhang, D., You, J.: Multi-object grasping detection with hierarchical feature fusion. IEEE Access 7, 43884–43894 (2019)
Gao, F., Wang, C., Li, C.: A combined object detection method with application to pedestrian detection. IEEE Access 8, 194457–194465 (2020)
Tan, P., Mao, K., Zhou, S.: Image target detection algorithm for smart city management cases. IEEE Access 8, 163357–163364 (2020)
Guo, S., Liu, F., Yuan, X., Zou, C., Chen, L., Shen, T.: HSPOG: an optimized target recognition method based on histogram spatial pyramid of oriented gradients. Tsinghua Sci. Technol. 26(4), 475–483 (2021)
Zhang, Z., Zou, C., Han, P., Lu, X.: A runway detection method based on classification using optimized polarimetric features and reliable re-detection scheme. IEEE Access 8, 49160–49168 (2020)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Mishra, S., Upadhyay, D., Saranya, P. (2022). Robust Vehicle Detection for Highway Monitoring Using Histogram of Oriented Gradients and Reduced Support Vector Machine. In: Singh, M., Tyagi, V., Gupta, P.K., Flusser, J., Ören, T. (eds) Advances in Computing and Data Sciences. ICACDS 2022. Communications in Computer and Information Science, vol 1613. Springer, Cham. https://doi.org/10.1007/978-3-031-12638-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-12638-3_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-12637-6
Online ISBN: 978-3-031-12638-3
eBook Packages: Computer ScienceComputer Science (R0)