Abstract

This paper proposes an improvement of Advanced Driver Assistance System based on saliency estimation of road signs. After a road sign detection stage, its saliency is estimated using a SVM learning. A model of visual saliency linking the size of an object and a size-independent saliency is proposed. An eye tracking experiment in context close to driving proves that this computational evaluation of the saliency fits well with human perception, and demonstrates the applicability of the proposed estimator for improved ADAS.

Abstract

This paper presents a 'CPN/B method' based process for railway systems safety analysis. Achieving interoperability through the European Rail Traffic Management System (ERTMS/ETCS) is facing difficulties in railway safety assessment due to the interaction of national and European operating specifications. These specifications have been modeled using several formalisms, which makes it is extremely hard to preserve all requirements when switching between different formalisms. However, this problem, crucial for efficient progress in railway safety research, has received very little attention in the literature. In this respect, the purpose of this contribution is to provide a methodology to demonstrate safety in railway systems by converting CPN models, widely used in modeling, into B abstract machines. It aims at enabling a stronger combination of formal design techniques and analysis tools able to cope with the real complexity of systems and automatically prove that safety properties are unambiguous, consistent and not contradictory, considering an industrial railway context. Document type: Article