Abstract
It is well-known that information and communication technologies enable many tasks in the context of precision agriculture. In fact, more and more farmers and food and agriculture companies are using precision agriculture-based systems to enhance not only their products themselves, but also their means of production. Consequently, problems arising from large amounts of data management and processing are arising. It would be very useful to have an infrastructure that allows information and agricultural tasks to be efficiently shared and handled. The cloud computing paradigm offers a solution. In this study, a cloud-based software architecture is proposed with the aim of enabling a complete crop management system to be deployed and validated. Such architecture includes modules developed by using Google App Engine, which allows the information to be easily retrieved and processed and agricultural tasks to be properly defined and planned. Additionally, Google's Datastore (which ensures a high scalability degree), hosts both information that describes such agricultural tasks and agronomic data. The architecture has been validated in a system that comprises a wireless sensor network with fixed nodes and a mobile node on an unmanned aerial vehicle (UAV), deployed in an agricultural farm in the Region of Murcia (Spain). Such a network allows soil water and plant status to be monitored. The UAV (capable of executing missions defined by an administrator) is useful for acquiring visual information in an autonomous manner (under operator supervision, if needed). The system performance has been analysed and results that demonstrate the benefits of using the proposed architecture are detailed.

Abstract
The detection of dissolved carbon dioxide (dCO(2)) is made possible through a colorimetric effect that occurs in a sensitive membrane. The reaction with dCO(2) changes the pH of the membrane causing a small difference in its colour which results in a characteristic absorbance spectrum band near 435 nm. A sensing platform based on this effect was developed and tested in gaseous and in aqueous environments. It is a combination of a bundle of large core fibre optics (with diameters above 200 mu m) with light emission diodes (LEDs) in the visible range of the spectrum, a silicon photodetector and a polymer membrane sensitive to CO2. A variation in the absorption of 3 / %VV was obtained in the range from 0 to 1.6 % of gaseous CO2 with an estimated response time below 60 seconds.

Abstract
It is in the nature of chaotic atmospheric processes that weather forecasts will never be perfectly accurate. This natural fact poses challenges not only for private life, public safety, and traffic but also for electrical power systems with high shares of weather-dependent wind and solar power production. © 2012 IEEE.

Abstract
Assessment of retinal vessels is fundamental for the diagnosis of many disorders such as heart diseases, diabetes and hypertension. The imaging of retina using advanced fundus camera has become a standard in computer-assisted diagnosis of opthalmic disorders. Modern cameras produce high quality color digital images, but during the acquisition process the light reflected by the retinal surface generates a luminosity and contrast variation. Irregular illumination can introduce severe distortions in the resulting images, decreasing the visibility of anatomical structures and consequently demoting the performance of the automated segmentation of these structures. In this paper, a novel approach for illumination correction of color fundus images is proposed and applied as preprocessing step for retinal vessel segmentation. Our method builds on the connection between two different phenomena, shadows and haze, and works by removing the haze from the image in the inverted intensity domain. This is shown to be equivalent to correct the nonuniform illumination in the original intensity domain. We tested the proposed method as preprocessing stage of two vessel segmentation methods, one unsupervised based on mathematical morphology, and one supervised based on deep learning Convolutional Neural Networks (CNN). Experiments were performed on the publicly available retinal image database DRIVE. Statistically significantly better vessel segmentation performance was achieved in both test cases when illumination correction was applied.

Abstract
A Fabry-Perot air bubble microcavity fabricated between a section of single mode fiber and a multimode fiber that requires only the use of a commercial fusion splicer is proposed. The study of the microcavities growth with the number of applied arcs is performed and several sensors are tested. The sensors are tested for lateral load measurements, and it is observed that there is dependence between the sensor dimensions and its sensitivity. The maximum sensitivity of 2.11 nm/N was obtained for the 161-mu m-long cavity. Moreover, given the low temperature sensitivity (<1 pm/degrees C), the proposed cavity should be adequate to perform temperature-independent measurements. The accurate technique control leads to the fabrication of reproducible cavities with the sensitivity required for the application. The way of manufacturing using a standard fusion splicer, given that no oils or etching solutions are involved, emerges as an alternative to the previously developed air bubble-based sensors.

Abstract
Pose invariant object detection and classification plays a critical role in robust image recognition systems and can be applied in a multitude of applications, ranging from simple monitoring to advanced tracking. This paper analyzes the usage of the Bag of Words model for recognizing objects in different scales, orientations and perspective views within cluttered environments. The recognition system relies on image analysis techniques, such as feature detection, description and clustering along with machine learning classifiers. For pinpointing the location of the target object, it is proposed a multiscale sliding window approach followed by a dynamic thresholding segmentation. The recognition system was tested with several configurations of feature detectors, descriptors and classifiers and achieved an accuracy of 87% when recognizing cars from an annotated dataset with 177 training images and 177 testing images. © Springer International Publishing AG 2018.