In this article, we examine the impacts of AI on intelligent image analysis applications for global security, which we have been specializing in at EVITECH since 2005 [1].

These image analysis applications aim at detecting dangerous situations to prevent or limit their consequences: intrusion in a sensitive site, dangerous individual behavior, detection of a hydrocarbon leak or smoke, crowd control (counting, flow, incidents), presence of suspicious or dangerous object, etc. After a general presentation of the history of the discipline, we identify the most relevant contributions of AI for video analysis, in theoretical terms, but also in practical terms (AI has facilitated analysis in public space), the capacities of video surveillance operators not being unlimited either in terms of energy consumed or in price. We cannot conclude without addressing the legal issue: the law limits these uses in public space.

[1] "Interest and limits of intelligent video surveillance for global security", P. Bernas, Conf. AVIRS 2008.

To access the full article, follow this link: White paper