Software development for smart cities bring into light new concerns, such as how to deal with scalability, heterogeneity (sensors, actuators, high performance computing devices, etc.), geolocation information or privacy issues, among some. Traditional approaches to distributed systems fail to address these challenges, because they were mainly devoted to enterprise IT environments. This paper proposes a middleware framework, called Civitas, specially devoted to support the task of service development for the Smart City paradigm. This paper also analyzes the main drawbacks of traditional approaches to middleware service development and how these are overcome in the middleware framework proposed here.
16 March 2024 in International Journal of Neural Systems
The benefits that technology can provide in terms of health and support for independent living are in many cases not enough to break the barriers that prevent older adults from accepting and embracing technology. This work proposes a hardware and software platform based on a smart mirror, which is equipped with a set of digital solutions whose main focus is to overcome older adults’ reluctance to use technology at home and wearable devices on the move. The system has been developed in the context of two use cases: the support of independent living for older individuals with neurodegenerative diseases and the promotion of physical rehabilitation activities at home. Aspects such as reliability, usability, consumption of computational resources, performance and accuracy of the proposed platform and digital solutions have been evaluated in the initial stages of the pilots within the SHAPES project, an EU-funded innovation action. It can be concluded that the SHAPES smart mirror has the potential to contribute as a technological breakthrough to overcome the barriers that prevent older adults from engaging in the use of assistive technologies.
This paper presents a novel method that leverages reasoning capabilities in a computer vision system dedicated to human action recognition. The proposed methodology is decomposed into two stages. First, a machine learning based algorithm - known as bag of words - gives a first estimate of action classification from video sequences, by performing an image feature analysis. Those results are afterward passed to a common-sense reasoning system, which analyses, selects and corrects the initial estimation yielded by the machine learning algorithm. This second stage resorts to the knowledge implicit in the rationality that motivates human behaviour. Experiments are performed in realistic conditions, where poor recognition rates by the machine learning techniques are significantly improved by the second stage in which common-sense knowledge and reasoning capabilities have been leveraged. This demonstrates the value of integrating common-sense capabilities into a computer vision pipeline.