Video-Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation-Matched Tracking Using Image Data from PTZ Cameras

Vehicle counting plays a significant role in vehicle behavior analysis and traffic incident detection for established video surveillance systems on expressway. Since the existing sensor method and the traditional image processing method have the problems of difficulty in installation, high cost, and low precision, a novel vehicle counting method is proposed, which realizes efficient counting based on multivehicle detection and multivehicle tracking. For multivehicle detection tasks, a construction of the new expressway dataset consists of a large number of sample images with a high resolution (1920 × 1080) captured from real-world expressway scenes (including the diversity climatic conditions and visual angles) by Pan-Tilt-Zoom (PTZ) cameras, in which vehicle categories and annotation rules are defined. Moreover, a correlation-matched algorithm for multivehicle tracking is proposed, which solves the problem of occlusion and vehicle scale change in the tracking process. Due to the discontinuity and unsmooth of the trajectories that occurred during the tracking process, we designed a trajectory optimization algorithm based on least square method. Finally, a new vehicle counting method is designed based on the tracking results, in which the driving direction information of the vehicle is added in the counting process. The experimental results show that the proposed counting method in this research can achieve more than 93% accuracy and an average speed of 25 frames per second in expressway video sequence.

[1]  Fei Liu,et al.  A video-based real-time adaptive vehicle-counting system for urban roads , 2017, PloS one.

[2]  A. Sumalee,et al.  Automatic Freeway Incident Detection for Free Flow Conditions: A Vehicle Reidentification Based Approach Using Image Data from Sparsely Distributed Video Cameras , 2015 .

[3]  Hyondong Oh,et al.  New Multiple-Target Tracking Strategy Using Domain Knowledge and Optimization , 2017, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[4]  Jiashu Zhang,et al.  A new real-time eye tracking based on nonlinear unscented Kalman filter for monitoring driver fatigue , 2010 .

[5]  Giovanni Maria Farinella,et al.  An integrated system for vehicle tracking and classification , 2015, Expert Syst. Appl..

[6]  Shiru Qu,et al.  Real-time vehicle detection and counting in complex traffic scenes using background subtraction model with low-rank decomposition , 2017 .

[7]  David Beymer,et al.  A real-time computer vision system for vehicle tracking and traffic surveillance , 1998 .

[8]  Thomas Brox,et al.  High Accuracy Optical Flow Estimation Based on a Theory for Warping , 2004, ECCV.

[9]  L. Javier García-Villalba,et al.  Vehicle Counting in Video Sequences: An Incremental Subspace Learning Approach , 2019, Sensors.

[10]  Wei Li,et al.  Background updating algorithm based on classification in complex scene: Background updating algorithm based on classification in complex scene , 2009 .

[11]  Zhong Chen,et al.  End-to-End Airplane Detection Using Transfer Learning in Remote Sensing Images , 2018, Remote. Sens..

[12]  Jian Sun,et al.  Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition , 2015, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  Yuanbo Guo,et al.  HeteMSD: A Big Data Analytics Framework for Targeted Cyber-Attacks Detection Using Heterogeneous Multisource Data , 2019, Secur. Commun. Networks.

[14]  M. A. Abdelwahab,et al.  Fast approach for efficient vehicle counting , 2019, Electronics Letters.

[15]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.