Abstract
The 4th industrial revolution advent promotes the reorganization of the traditional hierarchical automation systems towards decentralized Cyber-Physical Systems (CPS). In this context, Artificial Intelligence (AI) can address the new requirements through the use of data-driven and distributed problem solving approaches, such those based on Machine-Learning and Multi-agent Systems. Although their promising perspectives to enable and manage intelligent Internet of Things environments, the traditional Cloud-based AI approaches are not suitable to handle many industrial scenarios, constrained by responsiveness and data sensitive. The solution lies in taking advantage of Edge and Fog computing to create a decentralized multi-level data analysis computing infrastructure that supports the development of industrial CPS. However, this is not a straightforward task, posing several challenges and demanding new approaches and technologies. In this context, this work discusses the distribution of intelligence along Cloud, Fog and Edge computing layers in industrial CPS, leveraging some research challenges and future directions. Copyright

Abstract
Smart governance is one of the characteristics of smart cities, having its roots in e-government, in the principles of good governance, and in the assumptions of citizens' participation and involvement in public decision-making. This chapter aims to answer the question: “What smart governance practices are being implemented in smart cities” through an extensive literature review in the areas of e-government, good governance, smart cities and smart governance, and content analysis of the websites of seven smart cities: Amsterdam, Barcelona, Copenhagen, Lisbon, Manchester, Singapore, and Stockholm. The objective was to identify the presence of factors related with e-participation; e-services; and public administration functioning on the cities' websites. The chapter ends with directions for future research and the conclusion that all the smart cities analyzed presented some factors related with smart governance, but with different levels of development and application.

Abstract
Smart Governence as its roots in e-government, in the principles of good governance, and in the assumptions of citizens' participation and involvement in public decision-making and is considered one of the six main characteristics of smart cities. The present investigation was intended to answer the question: "What smart governance practices are being implemented in European smart cities" through an extensive literature review and content analysis of the websites of six European smart cities: Amsterdam; Barcelona; Copenhagen; Lisbon; Manchester and Stockholm. The objective was to identify the presence of factors related with e-participation; e-services; and the functioning of local public administration on the city's websites. It was concluded that all the smart cities analyzed presented some factors related with smart governance, but with different levels of development and application.

Abstract
We present PRIME (PSF Reconstruction and Identification for Multi-sources characterization Enhancement) as a novel hybrid concept to improve the PSF estimation based on Adaptive optics (AO) control loop data. PRIME uses both focal and pupil plane data to jointly estimate the model parameters, which are both the atmospheric (Cn2 (h), seeing), system (e.g. optical gains, residual low-order errors). The parametric model in use is flexible enough to be scaled with field location and wavelength, making it a proper choice for optimized on-axis and off-axis data-reduction across the spectrum. We review the methodology and on-sky validations on NIRC2 at Keck II. We also present applications of PSF model parameters retrieval using PRIME: (i) calibrate the PSF model for observations void of stars on the acquired images, i.e. optimize the PSF reconstruction process (ii) update the AO error breakdown mutually constrained by the telemetry and the images in order to speculate on the origin of the missing error terms and evaluate their magnitude (iii) measure photometry and astrometry in stellar fields.

Abstract
Frequently software testing tends to be neglected at the beginning of the projects, only performed on the late stage. However, it is possible to benefit from combining requirement with testing specification activities. On one hand, acceptance tests specification will require less manual effort since they are defined or generated automatically from the requirements specification. On the other hand, the requirements specification itself will end up having higher quality due to the use of a more structured language, reducing typical problems such as ambiguity, inconsistency and incorrectness. This research proposes an approach that promotes the practice of tests specification since the very beginning of projects, and its integration with the requirements specification itself. It is a model-driven approach that contributes to maintain the requirements and tests alignment, namely between requirements, test cases, and low-level automated test scripts. To show the applicability of the approach, two complementary languages are adopted: the ITLingo RSL that is particularly designed to support both requirements and tests specification; and the Robot language, which is a low-level keyword-based language for the specification of test scripts. The approach includes model-to-model transformation techniques, such as test cases into test scripts transformations. In addition, these test scripts are executed by the Robot test automation framework.

Abstract
The segmentation of retinal vessels in fundus images has been heavily focused in the past years, given their relevance in the diagnosis of several health conditions. Even though the recent advent of deep learning allowed to foster the performance of computer-based algorithms in this task, further improvement concerning the detection of vessels while suppressing background noise has clinical significance. Moreover, the best performing state-of-the-art methodologies conduct patch-based predictions. This, put together with the preprocessing techniques used in those methodologies, may hinder their use in screening scenarios. Thus, in this paper, we explore a fully convolutional setting that takes raw fundus images and allows to combine patch-based training with global image prediction. Our experiments on the DRIVE, STARE and CHASEDB1 databases show that the proposed methodology achieves state-of-the-art performance in the first and the last, allowing at the same time much faster segmentation of new images.