Article ID Journal Published Year Pages File Type
525842 Computer Vision and Image Understanding 2010 11 Pages PDF
Abstract

Intelligent video surveillance (IVS) technology is on the cusp of moving from early adopters to general acceptance in several markets such as security and business intelligence. This transition has been made possible by embedding computer vision technologies directly into video devices such as cameras, encoders, routers, DVRs, NVRs, and other video management and storage hardware. For this technology to be successful, it is crucial that IVS systems can be deployed easily, without requiring computer vision expertise to customize them for every installation; and that the systems work robustly in a wide range of environments. One of the key enablers to achieve this goal is proper testing. This paper discusses some of the major challenges involved and provides a case study for addressing the problem. One of the key concepts is utilizing fuzzy evaluation to handle boundary conditions.

Research highlights► Commercial intelligent video surveillance system testing methods. ► Hands-on experiences of evaluating IVS system. ► Fuzzy concepts and methods in testing and evaluating IVS system. ► A case study of performance evaluation.

Related Topics
Physical Sciences and Engineering Computer Science Computer Vision and Pattern Recognition
Authors
, ,