Accurate vehicle detection and counting algorithm for traffic data collection

Number of vehicles on road is very important traffic data and is essential for transportation safety and management. In this paper, an approach for vehicle detection is presented. In this approach, virtual line based sensors which are just straight detection lines are first set on road lanes. Then two features, namely gradient and range feature, are proposed for vehicle detection. This is carried out by extracting and analyzing the two features on detection lines. Meanwhile, the solution for vehicle occlusion have also been proposed. Our proposed method has an outstanding advantage that it performs excellent in traffic jams as well as under various conditions, such as sunny, cloudy, and rainy days, or night time, or even tunnels with complex illumination. The high accuracy rate of our method is verified with the experiment results.

[1]  Antonio Fernández-Caballero,et al.  Vehicle Tracking by Simultaneous Detection and Viewpoint Estimation , 2013, IWINAC.

[2]  Mau-Tsuen Yang,et al.  Traffic flow estimation and vehicle-type classification using vision-based spatial-temporal profile analysis , 2013, IET Comput. Vis..

[3]  Shih-Chieh Huang,et al.  A Vision-Based Vehicle Speed Warning System , 2012, 2012 9th International Conference on Ubiquitous Intelligence and Computing and 9th International Conference on Autonomic and Trusted Computing.

[4]  Stanley T. Birchfield,et al.  Real-Time Incremental Segmentation and Tracking of Vehicles at Low Camera Angles Using Stable Features , 2008, IEEE Transactions on Intelligent Transportation Systems.

[5]  Wei Zhang,et al.  Multilevel Framework to Detect and Handle Vehicle Occlusion , 2008, IEEE Transactions on Intelligent Transportation Systems.

[6]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Ghazali Sulong,et al.  Vehicle Detection and Tracking Techniques: A Concise Review , 2014, ArXiv.

[8]  Shuguang Li,et al.  Video-Based Traffic Data Collection System for Multiple Vehicle Types , 2012 .

[9]  Madasu Hanmandlu,et al.  Estimation of vehicle speed by motion tracking on image sequences , 2010, 2010 IEEE Intelligent Vehicles Symposium.

[10]  Edward Jones,et al.  Rear-Lamp Vehicle Detection and Tracking in Low-Exposure Color Video for Night Conditions , 2010, IEEE Transactions on Intelligent Transportation Systems.

[11]  Santosh Kumar,et al.  Vehicle Speed Detection Using Corner Detection , 2014, 2014 Fifth International Conference on Signal and Image Processing.

[12]  W. Jatmiko,et al.  Vehicle counting and speed measurement using headlight detection , 2013, 2013 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[13]  Farid Melgani,et al.  Automatic Car Counting Method for Unmanned Aerial Vehicle Images , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[14]  M. P. Sirisha,et al.  Adaptive Vehicle Detector Approach for Complex Environments , 2013 .

[15]  S. M. Mahbubur Rahman,et al.  Detection and Classification of Vehicles From Video Using Multiple Time-Spatial Images , 2012, IEEE Transactions on Intelligent Transportation Systems.

[16]  Mohan M. Trivedi,et al.  Looking at Vehicles on the Road: A Survey of Vision-Based Vehicle Detection, Tracking, and Behavior Analysis , 2013, IEEE Transactions on Intelligent Transportation Systems.

[17]  Xiaogang Wang,et al.  Counting Vehicles from Semantic Regions , 2013, IEEE Transactions on Intelligent Transportation Systems.

[18]  Jacob Scharcanski,et al.  Tracking and counting vehicles in traffic video sequences using particle filtering , 2013, 2013 IEEE International Instrumentation and Measurement Technology Conference (I2MTC).

[19]  Li Shuguang,et al.  Video-based traffic data collection system for multiple vehicle types , 2014 .