Intelligent surveillance refers to using Artificial Intelligence techniques in order to improve surveillance and deal with semantic information obtained from low-level security devices. In this context, the use of expert knowledge may offer a more realistic solution when dealing with the design of a surveillance system. In this work, a conceptual framework based on normality analysis to detect abnormal behaviors by means of normality concepts is presented. A normality concept specifies how a certain object should ideally behave in a concrete environment depending on such a concept. The definition of the normal path concept is studied in depth in order to analyze behaviors in an outdoor environment.

Highlights We design and develop an intelligent tutoring system to encourage students to learn through experimentation We allow to work at higher levels of Bloom’s Taxonomy with the designed intelligent tutoring system We use artificial intelligence and fuzzy rules to design and develop the learning instructor We establish foundations to build intelligent tutoring system that allow to learn through experimentation

Physical rehabilitation of stroke patients is based on the daily execution of exercises with face-to-face supervision by therapists. This model cannot be sustained in the long term, due to the involved economic costs, the growing number of patients, and the aging population. Remote rehabilitation tools have emerged to address this unmet clinical need, but they face the double challenge of motivating patients and ensuring an effective remote rehabilitation. In this context, exergames allow patients to play while performing repetitive therapeutic tasks in a safe and ecological environment. This work proposes the design of Personalized Exergames Language (PEL), a language whose sentences can be processed via software in order to automatically generate exergames. The definition of exergames through PEL, guided by an effective methodology of the design and generation of personalized exergames, will include both game mechanics and the necessary metrics to monitor, guide, and adapt the rehabilitation of each patient. The integration of authoring tools are considered to visually guide the therapist when designing exergames. A study has been carried out with stroke patients and therapists from a hospital and two community centers, in order to evaluate several exergames, automatically generated using PEL, in terms of usability, understanding, and suitability.

No abstract

Intelligent surveillance involves the use of AI techniques to monitor environments whose analysis is becoming more and more complex because of the large number of sensors used and the need of monitoring multiple events of interest simultaneously. Most of the current surveillance systems provide solutions for particular problems but still suffer from lack of flexibility and scalability when they are used on different or related surveillance problems. To overcome this limitation, two aspects should be addressed: a knowledge-based surveillance model flexible enough to deal with different events of interest and an architecture that gives support to this model when deploying the surveillance system within a particular scenario. This paper discusses the architecture devised to deploy intelligent surveillance systems by means of a set of autonomous agents that are responsible for the management of different surveillance tasks and for cooperating to monitor complex environments. This multi-agent architecture is inspired by a normality-based formal model used to define the knowledge needed to analyze general-purpose surveillance concepts. We use the architecture to deploy a surveillance system to monitor an urban traffic scenario.