Abstract
Monitoring the quality of high refractive index edible oils is of great importance for the human health. Uncooked edible oils in general are healthy foodstuff, olive oil in particular, however, they are frequently used for baking and cooking. High quality edible oils are made from seeds, nuts or fruits by mechanical processes. Nevertheless, once the mechanical extraction is complete, up to 15% of the oil remains in oil pomace and in the mill wastewater, which can be extracted using organic solvents, often hexane. Optical fiber sensors based on long period fiber gratings (LPFG) have very low wavelength sensitivity when the surround refractive index is higher than the refractive index of the cladding. Titanium dioxide (TiO2) coated LPFG could lead to the realization of high sensitivity chemical sensor for the food industry. In this work LPFG coated with a TiO2 thin film were successfully used for to detect small levels of hexane diluted in edible oils and for real time monitoring the thermal deterioration of edible oils. For a TiO2 coating of 30 nm a wavelength sensitivity of 1361.7 nm/RIU (or 0.97 nm /% V/V) in the 1.4610-1.4670 refractive index range was achieved, corresponding to 0 to 12 % V/V of hexane in olive oil. A sensitivity higher than 638 nm/RIU at 225 degrees C was calculated, in the 1.4670-1.4735 refractive index range with a detection limit of thermal deterioration of about 1 minute.

Abstract
Recognizing a place with a visual glance is the first capacity used by humans to understand where they are. Making this capacity available to robots will make it possible to increase the redundancy of the localization systems available in the robots, and improve semantic localization systems. However, to achieve this capacity it is necessary to build a robust visual signature that could be used by a classifier. This paper presents a new approach to extract a global descriptor from an image that can be used as the visual signature for indoor scenarios. This global descriptor was tested using videos acquired from three robots in three different indoor scenarios. This descriptor has shown good accuracy and computational performance when compared to other local and global descriptors.

Abstract
Robustness, real-time properties and resource efficiency are key properties to embedded devices of the CPS/IoT era. In this paper we propose a language approach RTFM-core, and show its potential to facilitate the development process and provide highly efficient and statically verifiable implementations. Our programming model is reactive, based on the familiar notions of concurrent tasks and (single-unit) resources. The language is kept minimalistic, capturing the static task, communication and resource structure of the system. Whereas C-source can be arbitrarily embedded in the model, and/or externally referenced, the instep to mainstream development is minimal, and a smooth transition of legacy code is possible. A prototype compiler implementation for RTFM-core is presented. The compiler generates C-code output that compiled together with the RTFM-kernel primitives runs on bare metal. The RTFM-kernel guarantees deadlock-lock free execution and efficiently exploits the underlying interrupt hardware for static priority scheduling and resource management under the Stack Resource Policy. This allows a plethora of well-known methods to static verification (response time analysis, stack memory analysis, etc.) to be readily applied. The proposed language and supporting tool-chain is demonstrated by showing the complete process from RTFM-core source code into bare metal executables for a lightweight ARM-Cortex M3 target.

Abstract
Fingerprint segmentation is a crucial step of an automatic fingerprint identification system, since an accurate segmentation promote both the elimination of spurious minutiae close to the foreground boundaries and the reduction of the computation time of the following steps. In this paper, a new, and more robust fingerprint segmentation algorithm is proposed. The main novelty is the introduction of a more robust binarization process in the framework, mainly based on the fuzzy C-means clustering algorithm. Experimental results demonstrate significant benchmark progress on three existing FVC datasets.

Abstract
Traditionally, travel demand modelling focused on long-term multiple socio-economic scenarios and land-use configurations to estimate the required transport supply. However, the limited number of transportation requests in demand-responsive flexible transport systems require a higher resolution zoning. This work analyses users short-term destination choice patterns, with a careful analysis of the available data coming from various different sources, such as GPS traces and social networks. We use a Multinomial Logit Model, with a social component for utility and characteristics, both derived from Social Network Analyses. The results from the model show meaningful relationships between distance and attractiveness for all the different alternatives, with the variable distance being the most significant.

Abstract
Intelligent wheelchairs (IW) are technologies that can increase the autonomy and independence of elderly people and patients suffering from some kind of disability. Nowadays the intelligent wheelchairs and the human-machine studies are very active research areas. This paper presents a methodology and a Data Analysis System (DAS) that provides an adapted command language to an user of the IW. This command language is a set of input sequences that can be created using inputs from an input device or a combination of the inputs available in a multimodal interface. The results show that there are statistical evidences to affirm that the mean of the evaluation of the DAS generated command language is higher than the mean of the evaluation of the command language recommended by the health specialist (p value = 0.002) with a sample of 11 cerebral palsy users. This work demonstrates that it is possible to adapt an intelligent wheelchair interface to the user even when the users present heterogeneous and severe physical constraints.