Abstract
Over the past few decades, much has been done globally to make final energy consumption more sustainable, integrating more renewable resources. In Europe, there are many directives and regulations that imposes to the states ambitious targets on the promotion of renewable energies and energetic efficiency among others. Portugal has not only fulfilled all the goals but has also exceeded them. This is the case with the integration of renew ables in electricity consumption, which is currently rated at 53 %, with the prospect of reaching 80% in 2030. In this scenario, it is important to encourage the energy production from the solar resource as the country has optimal conditions for this purpose. This research assess the existing Portuguese potential for floating photovoltaic systems and its integration in the power grid. Another aim of this study is to sizing and assesses the energetic potential of floating solar power plant in Gouvaes dam, included in the Tamega hydroelectric complex in under construction in northern Portugal belong to Iberdrola Company.

Abstract
Realizing autonomous inspection, such as that of power distribution lines, through unmanned aerial vehicle (UAV) systems is a key research domain in robotics. In particular, the use of autonomous and semi-autonomous vehicles to execute the tasks of an inspection process can enhance the efficacy and safety of the operation; however, many technical problems, such as those pertaining to the precise positioning and path following of the vehicles, robust obstacle detection, and intelligent control, must be addressed. In this study, an innovative architecture involving an unmanned aircraft vehicle (UAV) and an unmanned ground vehicle (UGV) was examined for detailed inspections of power lines. In the proposed strategy, each vehicle provides its position information to the other, which ensures a safe inspection process. The results of real-world experiments indicate a satisfactory performance, thereby demonstrating the feasibility of the proposed approach.

Abstract
In order to take advantage of the processing power of current computing platforms, programmers typically need to develop software versions for different target devices. This task is time-consuming and requires significant programming and computer architecture expertise. A possible and more convenient alternative is to start with a single high-level description of a program with minimum implementation details, and generate custom implementations according to the target platform. In this paper, we use MATLAB as a high-level programming language and propose a compiler that targets CPU/GPU computing platforms by generating customized implementations in C and OpenCL. We propose a number of compiler techniques to automatically generate efficient C and OpenCL code from MATLAB programs. One of such compiler techniques relies on heuristics to decide when and how to use Shared Virtual Memory (SVM). The experimental results show that our approach is able to generate code that provides significant speedups (eg, geometric mean speedup of 11x for a set of simple benchmarks) using a discrete GPU over equivalent sequential C code executing on a CPU. With more complex benchmarks, for which only some code regions can be parallelized, and are thus offloaded, the generated code achieved speedups of up to 2.2x. We also show the impact of using SVM, specifically fine-grained buffers, and the results show that the compiler is able to achieve significant speedups, both over the versions without SVM and with naive aggressive SVM use, across three CPU/GPU platforms.

Abstract
The interest of using light in clinical practice is increasing strongly and many applications work at various wavelengths from the ultraviolet to the infrared. Due to this great range of applications, the determination of the optical properties of biological tissues in a wide spectral range becomes of interest. The liver is an important organ, since it has a major role in the human body and various pathologies are known to develop within it. For these reasons, this study concerns the estimation of the optical properties of human normal and pathological (metastatic carcinoma) liver tissues between 200 and 1000 nm. The obtained optical properties present the expected wavelength dependencies for both tissues - the refractive index, the absorption and the scattering coefficients decrease with the wavelength and the anisotropy and light penetration depth increase with the wavelength. Although similar behavior was observed for the various properties between the normal and pathological tissues, evidence of smaller blood content in the pathological tissues was found. A possible explanation is that the cancer cells destroy liver's vasculature and internal architecture, providing though a reduction in the blood content. For low wavelengths, it was observed a matching between the scattering and the reduced scattering coefficients, which implies a nearly zero anisotropy in that range. The scattering coefficient decreases from nearly 140 cm(-1) (at 200 nm) to 80 cm(-1) (at 1000 nm) for the normal liver and from nearly 140 cm(-1) (at 200 nm) to 95 cm(-1) (at 1000 nm) for the pathological tissue.

Abstract

Abstract
Convolution Neural Network (CNN)-based object detection models have achieved unprecedented accuracy in challenging detection tasks. However, existing detection models (detection heads) trained on 8-bits/pixel/channel low dynamic range (LDR) images are unable to detect relevant objects under lighting conditions where a portion of the image is either under-exposed or over-exposed. Although this issue can be addressed by introducing High Dynamic Range (HDR) content and training existing detection heads on HDR content, there are several major challenges, such as the lack of real-life annotated HDR dataset(s) and extensive computational resources required for training and the hyper-parameter search. In this paper, we introduce an alternative backwards-compatible methodology to detect objects in challenging lighting conditions using existing CNN-based detection heads. This approach facilitates the use of HDR imaging without the immediate need for creating annotated HDR datasets and the associated expensive retraining procedure. The proposed approach uses HDR imaging to capture relevant details in high contrast scenarios. Subsequently, the scene dynamic range and wider colour gamut are compressed using HDR to LDR mapping techniques such that the salient highlight, shadow, and chroma details are preserved. The mapped LDR image can then be used by existing pre-trained models to extract relevant features required to detect objects in both the under-exposed and over-exposed regions of a scene. In addition, we also conduct an evaluation to study the feasibility of using existing HDR to LDR mapping techniques with existing detection heads trained on standard detection datasets such as PASCAL VOC and MSCOCO. Results show that the images obtained from the mapping techniques are suitable for object detection, and some of them can significantly outperform traditional LDR images.