Using Deep Learning for Collecting Data about Museum Visitor Behavior

Nowadays, technology makes it possible to admire objects and artworks exhibited all over the world remotely. We have been able to appreciate this convenience even more in the last period, in which the pandemic has forced us into our homes for a long time. However, visiting art sites in person remains a truly unique experience. Even during on-site visits, technology can help make them much more satisfactory, by assisting visitors during the fruition of cultural and artistic resources. To this aim, it is necessary to monitor the active user for acquiring information about their behavior. We, therefore, need systems able to monitor and analyze visitor behavior. The literature proposes several techniques for the timing and tracking of museum visitors. In this article, we propose a novel approach to indoor tracking that can represent a promising and non-expensive solution for some of the critical issues that remain. In particular, the system we propose relies on low-cost equipment (i.e., simple badges and off-the-shelf RGB cameras) and harnesses one of the most recent deep neural networks (i.e., Faster R-CNN) for detecting specific objects in an image or a video sequence with high accuracy. An experimental evaluation performed in a real scenario, namely, the “Exhibition of Fake Art” at Roma Tre University, allowed us to test our system on site. The collected data has proven to be accurate and helpful for gathering insightful information on visitor behavior.

Robust Attention Deraining Network for Synchronous Rain Streaks and Raindrops Removal

<div>Synchronous Rain streaks and Raindrops Removal (SR3) is a very hard and challenging task, since rain streaks and raindrops are two wildly divergent real-scenario phenomena with different optical properties and mathematical distributions. As such, most of existing deep learning-based Singe Image Deraining (SID) methods only focus on one of them or the other. To solve this issue, we propose a new, robust and hybrid SID model, termed Robust Attention Deraining Network (RadNet) with strong robustenss and generalztion ability. The robustness of RadNet has two implications :(1) it can restore different degenerations, including raindrops, rain streaks, or both; (2) it can adapt to different data strategies, including single-type, superimposed-type and blended-type. Specifically, we first design a lightweight robust attention module (RAM) with a universal attention mechanism for coarse rain removal, and then present a new deep refining module (DRM) with multi-scales blocks for precise rain removal. The whole process is unified in a network to ensure sufficient robustness and strong generalization ability. We measure the performance of several SID methods on the SR3 task under a variety of data strategies, and extensive experiments demonstrate that our RadNet can outperform other state-of-the-art SID methods.</div>

Robust Attention Deraining Network for Synchronous Rain Streaks and Raindrops Removal

<div>Synchronous Rain streaks and Raindrops Removal (SR3) is a very hard and challenging task, since rain streaks and raindrops are two wildly divergent real-scenario phenomena with different optical properties and mathematical distributions. As such, most of existing deep learning-based Singe Image Deraining (SID) methods only focus on one of them or the other. To solve this issue, we propose a new, robust and hybrid SID model, termed Robust Attention Deraining Network (RadNet) with strong robustenss and generalztion ability. The robustness of RadNet has two implications :(1) it can restore different degenerations, including raindrops, rain streaks, or both; (2) it can adapt to different data strategies, including single-type, superimposed-type and blended-type. Specifically, we first design a lightweight robust attention module (RAM) with a universal attention mechanism for coarse rain removal, and then present a new deep refining module (DRM) with multi-scales blocks for precise rain removal. The whole process is unified in a network to ensure sufficient robustness and strong generalization ability. We measure the performance of several SID methods on the SR3 task under a variety of data strategies, and extensive experiments demonstrate that our RadNet can outperform other state-of-the-art SID methods.</div>

Song Richness Index: A measure to understand the diversity and repetition of notes in a birdsong

Birdsong plays an important role in mate attraction and territorial defense. Many birds, especially Passerines, produce varying sequences of multiple notes resulting in complex songs. Studying the diversity of notes within these songs can give insights into an individuals reproductive fitness. We first looked at the previously described and commonly used diversity measures to understand the possible case-specific limitations. We then developed a new diversity measure- Song Richness Index (SRI). We compared SRI with three measures of diversity using all possible combinations of notes to understand the case-specific advantages and limitations of all approaches. Simulating all possible combinations gave us insights into how each diversity measure works in a real scenario. SRI showed an advantage over conventional measures of diversity like Note Diversity Index (NDI), Shannons Equitability (SH), and Simpsons Diversity (SI), especially in the cases where songs are made up of only one type of repetitive note.

Bumbometer digital crowd game: collaboration through competition in entertainment events

The use of mobile devices, especially smartphones, is widespread across all social strata and age groups, helping to ensure faster access from anywhere, data collection, and more regular and frequent control to aid urban, environmental, and social management. In this scenario, the entertainment industry has benefited from this powerful individual technological resource in cultural and sporting events. In this way, this work presents a proposal for interaction and engagement in entertainment events in a more prosperous and more technological way, through the development of a collaborative and competitive mobile-­web crowd game, intended for enhancing interaction between the crowd and as a unified group, whether physically co-­located or online. The application, called Bumbometer, uses motion sensors during an interactive dynamic with the crowd applying concepts from Mobile Crowd Sensing and User eXperience. We conducted two experimental studies to evaluate the proposed technology, the first in a real scenario of a folk cultural festival and the second in a controlled environment, simulating an event considering a scenario in which users were geographically distant. The results indicate that people feel immersed and engaged during the interaction through the proposed game, which reinforces the statement that the game meets an increasingly growing need to use technologies to ensure more significant interaction and audience immersion at crowd entertainment events, a creative and far­-reaching form.

Case study: risk associated to wearing silver or graphene nanoparticle-coated facemasks for protection against COVID-19

The world is living a pandemic situation derived from the worldwide spreading of SARS-CoV-2 virus causing COVID-19. Facemasks have proven to be one of the most effective prophylactic measures to avoid the infection that has made that wearing of facemasks has become mandatory in most of the developed countries. Silver and graphene nanoparticles have proven to have antimicrobial properties and are used as coating of these facemasks to increase the effectivity of the textile fibres. In the case of silver nanoparticles, we have estimated that in a real scenario the systemic (internal) exposure derived from wearing these silver nanoparticle facemasks would be between 7.0 × 10–5 and 2.8 × 10–4 mg/kg bw/day. In addition, we estimated conservative systemic no effect levels between 0.075 and 0.01 mg/kg bw/day. Therefore, we estimate that the chronic exposure to silver nanoparticles derived form facemasks wearing is safe. In the case of graphene, we detected important gaps in the database, especially regarding toxicokinetics, which prevents the derivation of a systemic no effect level. Nevertheless, the qualitative approach suggests that the risk of dermal repeated exposure to graphene is very low, or even negligible. We estimated that for both nanomaterials, the risk of skin sensitisation and genotoxicity is also negligible.

The Usefulness of Video Learning Analytics in Small Scale E-Learning Scenarios

A variety of tools are available to collect, process and analyse learning data obtained from the clickstream generated by students watching learning resources in video format. There is also some literature on the uses of such data in order to better understand and improve the teaching-learning process. Most of the literature focuses on large scale learning scenarios, such as MOOCs, where videos are watched hundreds or thousands of times. We have developed a solution to collect clickstream analytics data applicable to smaller scenarios, much more common in primary, secondary and higher education, where videos are watched tens or hundreds of times, and to analyse whether the solution is useful to teachers to improve the learning process. We have deployed it in a real scenario and collected real data. Furthermore, we have processed and presented the data visually to teachers for those scenarios and have collected and analysed their perception of their usefulness. We conclude that the collected data are perceived as useful by teachers to improve the teaching and learning process.

A Thermal Discomfort Index for Demand Response Control in Residential Water Heaters

Demand-response techniques are crucial for providing a proper quality of service under the paradigm of smart electricity grids. However, control strategies may perturb and cause discomfort to clients. This article proposes a methodology for defining an index to estimate the discomfort associated with an active demand management consisting of the interruption of domestic electric water heaters. Methods are applied to build the index include pattern detection for estimating the water utilization using an Extra Trees ensemble learning method and a linear model for water temperature, both based on analysis of real data. In turn, Monte Carlo simulations are applied to calculate the defined index. The proposed approach is evaluated over one real scenario and two simulated scenarios to validate that the thermal discomfort index correctly models the impact on temperature. The simulated scenarios consider a number of households using water heaters to analyze and compare the thermal discomfort index for different interruptions and the effect of using different penalty terms for deviations of the comfort temperature. The obtained results allow designing a proper management strategy to fairly decide which water heaters should be interrupted to guarantee the lower discomfort of users.

ModelSet: a dataset for machine learning in model-driven engineering

The application of machine learning (ML) algorithms to address problems related to model-driven engineering (MDE) is currently hindered by the lack of curated datasets of software models. There are several reasons for this, including the lack of large collections of good quality models, the difficulty to label models due to the required domain expertise, and the relative immaturity of the application of ML to MDE. In this work, we present ModelSet, a labelled dataset of software models intended to enable the application of ML to address software modelling problems. To create it we have devised a method designed to facilitate the exploration and labelling of model datasets by interactively grouping similar models using off-the-shelf technologies like a search engine. We have built an Eclipse plug-in to support the labelling process, which we have used to label 5,466 Ecore meta-models and 5,120 UML models with its category as the main label plus additional secondary labels of interest. We have evaluated the ability of our labelling method to create meaningful groups of models in order to speed up the process, improving the effectiveness of classical clustering methods. We showcase the usefulness of the dataset by applying it in a real scenario: enhancing the MAR search engine. We use ModelSet to train models able to infer useful metadata to navigate search results. The dataset and the tooling are available at https://figshare.com/s/5a6c02fa8ed20782935c and a live version at http://modelset.github.io.

A Case Study for the Development and Use of a Well Control Simulator as a Digital Twin of a Real Scenario

Simulators in the petroleum industry have been used mainly for training purposes even though they present different applications like digital twins. In this regard, a simulator must approximate the well environment to reflect operative actions and reactions. This paper describes a case study where a well control simulator has been developed to be used as a digital twin where operators may try different scenarios in a safe environment before applying them to the physical well. To cover all aspects of the operation, the simulator should simulate surface equipment as well as a downhole environment. Numerical modeling techniques and hydraulic simulators are used to design the well response to operations. Different scenarios were established to cover most of the possible downhole problems and equipment malfunctions including electrical and hydraulic failures. The study compares a pre-determined set of KPIs common to three different types of simulation: well control, procedural and an integration of both. The target of the study is to collect the data resulting from the use of the simulator while it replicates a real-life situation. This virtual model of the rig and the well can be used to calibrate the main drilling parameters like SPM, RPM and WOB. The digital twin is also used to optimize operational procedures and improve performance and efficiency of rig crews as well as reduce their response time to possible problems. The results show an increase in performance when the knowledge of the rig is combined with the downhole feedback experience. This proves that training of the crew by reproducing their own equipment allows for a major jump in readiness and faster response with minimal mistakes. In addition, conducting the operation virtually allows the crew to uncover any possible issues before tackling the physical well. This in turn helps to reduce errors and safeguard both well and equipment integrity. This paper discusses the integration of the use of downhole environment behavior into a complete digital twin which will play an important role for providing a source of data for regular case studies concerning well control, Maintenance, Scheduling and other critical decisions. This new method candidates itself as a major contender for the future of simulation in the drilling business and shows the importance of that for reducing risks and errors.