Graph-cut based regional risk estimation for traffic scene

In this study, we investigate the regional risk estimation of drivers for street environment involving different players such as pedestrians, other vehicles, traffic signs, traffic lights, and crosswalks. Various researches focusing on objects regarding traffic have been realized by means of traditional risk estimation. In turn, conventional methods have not presented a realistic solution for drivers at risky regions and moments; whereas, our approach considers emerging risks for a driver due to dynamic actions of street players. A chessboard is devised for representing the street players, each of which carries different potential risks. Every square of the chessboard refers to a partition, which can host one or multiple players. Further, a partition can have different risks for a driver. The proposed model is realized using a graph-cut algorithm for energy minimization. Each partition is considered as a vertex of the graph, which can transfer risks caused by street players. Vertexes are formed via behavior as those of memory cell structures. The memory cells have risk transfer capabilities allowing a driver to determine momentarily risks on urban traffic. Consequently, this captures the regional risk for driver in light of the detected street players as demonstrated through the paper.

[1]  Chih-Ming Hsu,et al.  Detecting drivable space in traffic scene understanding , 2012, 2012 International Conference on System Science and Engineering (ICSSE).

[2]  R. Prevete,et al.  Color cues for traffic scene analysis , 1995, Proceedings of the Intelligent Vehicles '95. Symposium.

[3]  Marie-Pierre Jolly,et al.  Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[4]  R. Cucchiara,et al.  Statistic and knowledge-based moving object detection in traffic scenes , 2000, ITSC2000. 2000 IEEE Intelligent Transportation Systems. Proceedings (Cat. No.00TH8493).

[5]  Martin Lauer,et al.  3D Traffic Scene Understanding From Movable Platforms , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Gareth Funka-Lea,et al.  Graph Cuts and Efficient N-D Image Segmentation , 2006, International Journal of Computer Vision.

[7]  H. Eren,et al.  Similar association set based attribute selection algorithm for road profile detection , 2013, 2013 6th International Congress on Image and Signal Processing (CISP).

[8]  Yan Liu,et al.  Traffic scenes invariant vehicle detection , 2013, 2013 9th Asian Control Conference (ASCC).

[9]  Emilio Frazzoli,et al.  Multiple vehicle driving control for traffic flow efficiency , 2012, 2012 IEEE Intelligent Vehicles Symposium.

[10]  Mehmet Celenk,et al.  Interactive risky behavior model for 3-car overtaking scenario using joint Bayesian network , 2013, 2013 IEEE Intelligent Vehicles Symposium (IV).

[11]  Elwood S. Buffa,et al.  Graph Theory with Applications , 1977 .

[12]  Patrick Bouthemy,et al.  Motion segmentation and qualitative dynamic scene analysis from an image sequence , 1993, International Journal of Computer Vision.

[13]  Mohan M. Trivedi,et al.  Shadow detection algorithms for traffic flow analysis: a comparative study , 2001, ITSC 2001. 2001 IEEE Intelligent Transportation Systems. Proceedings (Cat. No.01TH8585).

[14]  Mehmet Celenk,et al.  Approaching car detection via clustering of vertical-horizontal line scanning optical edge flow , 2012, 2012 15th International IEEE Conference on Intelligent Transportation Systems.

[15]  Marie-Pierre Jolly,et al.  Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images , 2001, ICCV.

[16]  Luc Van Gool,et al.  Dynamic 3D Scene Analysis from a Moving Vehicle , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[17]  Paola Mello,et al.  Image analysis and rule-based reasoning for a traffic monitoring system , 2000, IEEE Trans. Intell. Transp. Syst..

[18]  H. Eren,et al.  An effective variable selection algorithm for Aggressive/Calm Driving detection via CAN bus , 2013, 2013 International Conference on Connected Vehicles and Expo (ICCVE).

[19]  Olga Veksler,et al.  Fast approximate energy minimization via graph cuts , 2001, Proceedings of the Seventh IEEE International Conference on Computer Vision.