Traffic congestion analysis: A new Perspective

In this paper, a new perspective of congestion is presented to promote the development of traffic video analysis. Our main contributions are threefold: a) An unified and quantifiable definition of congestion is proposed to describe the traffic state in video. b) Based on the definition, a congestion dataset which contains multiple traffic scenes is constructed as a platform for the research community. At the same time, a precise labeling method is introduced to get the ground truth of congestion level accurately. c) An algorithm based on Inverse Perspective Mapping (IPM) and pairwise regression is proposed to analyze traffic videos and serves as a baseline. We further compare the proposed method with two deep learning methods. Intensive experiments justify the effectiveness of the proposed method.

[1]  Jun Huang,et al.  An improved local binary pattern operator for texture classification , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[2]  Richard P. Wildes,et al.  Classification of traffic video based on a spatiotemporal orientation analysis , 2011, 2011 IEEE Workshop on Applications of Computer Vision (WACV).

[3]  A. E. Hoerl,et al.  Ridge regression: biased estimation for nonorthogonal problems , 2000 .

[4]  Ge Li,et al.  Robust lane marking detection using boundary-based inverse perspective mapping , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[5]  F. Hall.  TRAFFIC STREAM CHARACTERISTICS , 1997 .

[6]  D. S. Guru,et al.  Symbolic Classification of Traffic Video Shots , 2013 .

[7]  Shang-Hong Lai,et al.  Adaptive Foreground Object Extraction for Real-Time Video Surveillance with Lighting Variations , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[8]  Mahsa Ghasembaglou,et al.  HaarWavelet Based Distributed Predictive Target Tracking Algorithm for Wireless Sensor Networks , 2013 .

[9]  Chih-Hsien Hsia,et al.  Hierarchical Method for Foreground Detection Using Codebook Model , 2011, IEEE Trans. Circuits Syst. Video Technol..

[10]  Leizer Schnitman,et al.  Highway Traffic Congestion Classification using Holistic Properties , 2013 .

[11]  Shoab A. Khan,et al.  Traffic congestion classification using motion vector statistical features , 2013, Other Conferences.

[12]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[13]  Yuan Yuan,et al.  Congested scene classification via efficient unsupervised feature learning and density estimation , 2016, Pattern Recognit..

[14]  Juan José Vinagre-Díaz,et al.  Bluetooth Traffic Monitoring Systems for Travel Time Estimation on Freeways , 2016, IEEE Transactions on Intelligent Transportation Systems.

[15]  Carlo Tomasi,et al.  Good features to track , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[16]  Javier Sánchez,et al.  Regularization Strategies for Discontinuity-Preserving Optical Flow Methods , 2016, IEEE Transactions on Image Processing.

[17]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.

[18]  Matti Pietikäinen,et al.  Local Binary Patterns , 2010, Scholarpedia.

[19]  Vassilios Morellas,et al.  Robust Foreground Detection In Video Using Pixel Layers , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Yu Liu,et al.  Fast anomaly detection in traffic surveillance video based on robust sparse optical flow , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).