**1. Introduction**

56 Recent Developments in Video Surveillance

Yuk, J. S. C., K. Y. K. Wong, et al. (2007). Object-Based Surveillance Video Retrieval System

Zhang, C., X. Chen, et al. (2009). "Semantic retrieval of events from indoor surveillance video

databases." *Pattern Recognition Letters* 30(12): 1067-1076.

*and Recognition (ICIAR'07)*: 626-637.

with Real-Time Indexing Methodology. *International Conference on Image Analysis* 

Anyone who has experienced artifacts or freezing play while watching a film or a live sporting event on TV is familiar with the frustration accompanying sudden quality degradation at a key moment. Notwithstanding, video services with blurred images may have far more severe consequences for video surveillance practitioners. Therefore, the Quality of Experience (QoE) concept for video content used for entertainment differs considerably from the quality of video used for recognition tasks. This is because in the latter case subjective user satisfaction depends only or almost only on the possibility of achieving a given functionality (event detection, object recognition). Additionally, the quality of video used by a human observer for recognitions tasks is considerably different from objective video quality used in computer processing (Computer Vision).

So called, task-based videos require a special framework appropriate to the video's function — i.e. its use for recognition tasks rather than entertainment. Once the framework is in place, methods should be developed to measure the usefulness of the reduced video quality rather than its entertainment value. The precisely computed usefulness can be used to optimize not only the video quality but the whole surveillance system. It is especially important since surveillance systems often aggregates large number of cameras which streams has to be saved for possible future investigation. For example in Chicago at least 10,000 surveillance cameras are connected to a common storing system (ACLU, 2011).

To develop accurate objective measurements (models) for video quality, subjective experiments must be performed. For this purpose, the ITU-T1 P.910 Recommendation "Subjective video quality assessment methods for multimedia applications" (ITU-T, 1999) addresses the methodology for performing subjective tests in a rigorous manner.

In this chapter the methodology for performing subjective tests is presented. It is shown that subjective experiments can be very helpful nevertheless they have to be correctly prepared and analyzed. For illustration of the problem, license plates recognition analysis is shown in detail.
