**9. Acknowledgments**

This work was supported by the European Commission under the Grant INDECT No. FP7-218086.

#### **10. References**

ACLU (2011). Chicago's video surveillance cameras, *Technical report*, ACLU of Illinois.


is possible through the application of appropriate transformations, allowing the fitting of

The methodologies outlined in the chapter are just a single contribution to the overall framework of quality standards for task-based video. It is necessary to define requirements starting from the camera, through the broadcast, and until after the presentation. These

So far, the practical value of contribution is limited. It refers to limited scenarios. The presented approach is just a beginning of more advanced interesting framework of objective

Further steps have been planned in standardization in assessing task-based video quality with relation to QART initiative. Stakeholders list has been initially agreed and action points have been agreed. The plans include: quantifying VQiPS' GUCs, extending test methods (standardization of test methods and experimental designs of ITU-T P.912 Recommendation), measuring camera quality, investigating H.264 encoders, investigating video acuity as well as checking results' stability. The final outcome should be to share ideas on conducting joint experiments and publishing joint papers/VQEG reports, and finally, to submit joint standardization contributions. Plans/next steps for standardizing test methods and experimental designs include verification of issues like: subliminal cues, Computer-Generated Imagery (CGI) source video sequences, automated eye charts as well as subjects' proficiency. The agreed tasks include verifying requirements, refining methods/designs and, finally,

This work was supported by the European Commission under the Grant INDECT No.

Agresti, A. & Coull, B. A. (1998). Approximate is better than "exact" for interval estimation of binomial proportions, *The American Statistician* 52(2): 119–126. ISSN: 0003-1305.

Boone, W. J., Townsend, J. S. & Staver, J. (2010). Using Rasch theory to guide the practice

Duplaga, M., Leszczuk, M., Papir, Z. & Przelaskowski, A. (2008). Evaluation of

Eskicioglu, A. M. & Fisher, P. S. (1995). Image quality measures and their performance,

of survey development and survey data analysis in science education and to inform science reform efforts: An exemplar utilizing STEBI self-efficacy data, *Science*

quality retaining diagnostic credibility for surgery video recordings, *Proceedings of the 10th international conference on Visual Information Systems: Web-Based Visual Information Search and Management*, VISUAL '08, Springer-Verlag, Berlin, Heidelberg,

ACLU (2011). Chicago's video surveillance cameras, *Technical report*, ACLU of Illinois.

diverse recognition tasks into a single quality framework.

requirements will depend on scenario recognition.

quality assessment, described below (Leszczuk, 2011).

making subjective experiments both more accurate and feasible.

Agresti, A. (2002.). *Categorical Data Analysis*, 2nd edn, Wiley.

*Education* n/a: 1–23.

pp. 227–230.

Baker, F. B. (1985). *The Basics of Item Response Theory*, Heinemann.

*Communications, IEEE Transactions on* 43(12): 2959–2965. URL: http://dx.doi.org/10.1109/26.477498

**8. Future work**

**9. Acknowledgments**

FP7-218086.

**10. References**


URL: http://dx.doi.org/10.1007/978-3-642-13789-1\_4


URL: http://dx.doi.org/10.1007/978-3-642-21512-4\_2

Nyman, G., Radun, J., Leisti, T., Oja, J., Ojanen, H., Olives, J. L., Vuori, T. & Hakkinen, J. (2006). What do users really perceive — probing the subjective image quality experience, *Proceedings of the SPIE International Symposium on Electronic Imaging 2006: Imaging Quality and System Performance III, Vol. 6059*, pp. 1–7.

**1. Introduction**

Considered as a weak point in road and railway infrastructure, level crossings (LC) improvement safety became an important field of academic research and took increasingly railways undertakings concerns. Improving safety of persons and road-rail facilities is an essential key element to ensure a good operating of the road and railway transport. Statistically, nearly 44% of level crossings users have a bad perception of the environment which consequently increases the accidents risks Nelson (2002). However, the behavior of pedestrians, road vehicle drivers and railway operators cannot be previously estimated beforehand. According to Griffioen (2004), the human errors are the causes of 99% of accidents at LC whose 93% are caused by road users. It is important also to note the high cost related to each accident, approximately one hundred million euro per year in the EU for all level crossing accidents. For this purpose, road and railway safety professionals from several countries have been focused on providing a level crossings as safer as possible. Actions are planned in order to exchange information and provide experiments for improving the management of level crossing safety and performance. This has enabled us to discuss sharing knowledge gained

**Intelligent Surveillance System Based on Stereo** 

**Vision for Level Crossings Safety Applications** 

Nizar Fakhfakh, Louahdi Khoudour,

*Development and Networks (IFSTTAR)* 

*France* 

**5**

Jean-Luc Bruyelle and El-Miloudi El-Koursi *French Institute of Science and Technology for Transport,* 

High safety requirements for level crossing systems mean a high cost which hinders the technological setup of advanced systems. High technology systems are exploited and introduced in order to timely prevent collisions between trains and automobiles and to help reduce levels of risk from railroad crossings. Several conventional object detection systems have been tested on railroad crossings. These techniques provide more or less significant information accuracy. Any proposed system based on a technological solution is not intended to replace the present equipment installed on each level crossing. The purpose of such an intelligent system is to provide additional information to the human operator; it can be considered as support system operations. This concerns the detection and localization of any kind of objects, such as pedestrians, people on two-wheeled vehicle, wheelchairs and car drivers on the dangerous zone Yoda et al. (2006). Today, there are a number of trigger technologies installed at level crossings, but they all serve the same purpose: they detect moving object when passing at particular points in the LC. Indeed, those conventional obstacle detection systems have been used to prevent collisions between trains

from research into improving safety at level crossings.


URL: http://doi.acm.org/10.1145/1278760.1278762

