1,721,041 research outputs found
Mobile mpox detection system Supplementary Material
No full textThis repository contains the supplementary material accompanying the paper named: A Transfer Learning and Explainable Solution to Detect mpox from Smartphones images. Specifically, here can be found all the scripts to replicate the experiments and figures/tables presented in the manuscript.
Please, refer to the README.md file for more details.This work was produced with the co-funding European Union - Next Generation EU, in the context of The National Recovery and Resilience Plan. The funding derives partially from Investment 1.5 Ecosystems of Innovation, Project Tuscany Health Ecosystem (THE), CUP: B83C22003920001 in which the authors M. G. Campana and F. Delmastro are involved, from Project MUSA – Multilayered Urban Sustainability Action in the Investment 1.5 Ecosystems of Innovation in which the author S. Mascetti is involved, and from the Research and Innovation Program PE00000014, ``SEcurity and RIghts in the CyberSpace (SERICS)'', CUP: J33C22002810001, in which the author E. Pagani is involved
Mpox Close Skin Images
No full text<p>The <strong>Mpox Close Skin Images</strong> dataset (<strong>MCSI</strong>) is a collection of skin images obtained from diverse public sources, that we accurately pre-processed (i.e., cropped and zoomed) in order to focus the skin lesion (if present), and to evaluate Machine Learning models aimed at detecting different pathologies from skin lesion pictures taken with smartphone cameras.<br>
<br>
It includes a total of 400 pictures homogeneously divided in 4 different classes: <em>mpox</em>, which contains samples of mpox (formerly Monkeypox) skin lesions; <em>chickenpox</em>, with samples of chickenpox cases; <em>acne</em>, containing samples of acne at different severity levels; and <em>healthy</em>, which contains samples of skin without any evident symptoms.<br>
<br>
This repository is part of the supplementary material accompanying the paper named: <em>A Transfer Learning and Explainable Solution to Detect mpox from Smartphones images</em>.<br>
<br>
Please, refer to the <em>README.md</em> file for more details.</p>This work was produced with the co-funding European Union - Next Generation EU, in the context of The National Recovery and Resilience Plan. The funding derives partially from Investment 1.5 Ecosystems of Innovation, Project Tuscany Health Ecosystem (THE), CUP: B83C22003920001 in which the authors M. G. Campana and F. Delmastro are involved, from Project MUSA – Multilayered Urban Sustainability Action in the Investment 1.5 Ecosystems of Innovation in which the author S. Mascetti is involved, and from the Research and Innovation Program PE00000014, "SEcurity and RIghts in the CyberSpace (SERICS)", CUP J33C22002810001, in which the author E. Pagani is involved
SafeTrekker: towards automatic recognition of critical situations in mountain excursions
No full textWe present preliminary work on the design of a ubiquitous service that could support the participants of outdoor excursions, like trekkers, climbers or ski-mountaineers in identifying potentially dangerous situations as well as to alert with useful information the rescue teams in case of critical conditions. The system is based on a form of privacy preserving crowdsourcing: precise location and sensor-based information is acquired real-time by a trusted system component and can be used in case of emergency, while an anonymized dataset of each excursion is processed by a central component to create a model of "normal" situations. Together with a model of "potentially dangerous" situations created with specific domain knowledge, the system offers a decision support service that can also operate offline on the mobile device of the users
SafeBox : adaptable spatio-temporal generalization for location privacy protection
Full text linkSpatial and temporal generalization emerged in the literature as a common approach to preserve location privacy. However, existing solutions have two main shortcomings. First, spatiotemporal generalization can be used with different objectives: for example, to guarantee anonymity or to decrease the sensitivity of the location information. Hence, the strategy used to compute the generalization can follow different semantics often depending on the privacy threat, while most of the existing solutions are specifically designed for a single semantics. Second, existing techniques prevent the so-called inversion attack by adopting a top-down strategy that needs to acquire a large amount of information. This may not be feasible when this information is dynamic (e.g., position or properties of objects) and needs to be acquired from external services (e.g., Google Maps).
In this contribution we present a formal model of the problem that is compatible with most of the semantics proposed so far in the literature, and that supports new semantics as well. Our BottomUp algorithm for spatio-temporal generalization is compatible with the use of online services, it supports generalizations based on arbitrary semantics, and it is safe with respect to the inversion attack. By considering two datasets and two examples of semantics, we experimentally compare BottomUp with a more classical top-down algorithm, showing that BottomUp is efficient and guarantees better performance in terms of the average size (space and time) of the generalized regions
Accessible Mathematics on Touchscreen Devices: New Opportunities for People with Visual Impairments
Full text linkIn recent years educational applications for touchscreen devices (e.g., tablets) become widespread all over the world. While these devices are accessible to people with visual impairments, educational applications to support learning of STEM subjects are often not accessible to visually impaired people due to inaccessible graphics.
This contribution addresses the problem of conveying graphics to visual impaired users. Two approaches are taken into account: audio icons and image sonification. In order to evaluate the applicability of these approaches, we report our experience in the development of two didactic applications for touchscreen devices, specifically designed to support people with visual impairments or blindness while studying STEM subjects: Math Melodies and Audio Functions. The former is a commercial application to support children in primary school in an inclusive class. It adopts an interaction paradigm based on audio icons. The latter is a prototype application aimed at enabling visually impaired students to explore function diagrams and adopts an image sonification approach
An Efficient Algorithm for Minimizing Time Granularity Periodical Representations
No full textThis paper addresses the technical problem of efficiently reducing the periodic representation of a time granularity to its minimal form. The minimization algorithm presented in the paper has an immediate practical application: it allows users to intuitively define granularities (and more generally, recurring events) with algebraic expressions that are then internally translated to mathematical characterizations in terms of minimal periodic sets. Minimality plays a crucial role, since the value of the recurring period has been shown to dominate the complexity when processing periodic sets
Privacy Protection through Anonymity in Location-based Services
No full textThe adoption of location-based services (LBS) brings new privacy threats to users. The user location information revealed in LBS requests may be used by attackers to associate sensitive information of the user with her identity. This contribution focuses on privacy protection through anonymity, i.e., keeping individual users indistinguishable
in a large group of people that may have issued the same request. The contribution identifies different privacy threats to LBS users, discusses techniques for protecting user privacy under different threats, and gives a performance evaluation of the mentioned protection methods
Mapping Calendar Expressions into Periodical Granularities
Full text linkAn effort has been devoted in the recent years to study and formalize the concept of time granularity and to design applications and services using the formalization. Among
other proposals, a calendar algebra has been defined to facilitate the specification of new granularities and to perform conversions among them. This paper shows how granularities defined as algebraic calendar expressions can be represented as periodical sets of instants. More precisely, the
paper shows how each algebraic operator changes the periodical structure of the granularities given as operands. These results have an immediate application enabling users to easily specify new granularities and using them in the
only constraint solver supporting time granularities that is currently available
ZebraRecognizer : Pedestrian crossing recognition for people with visual impairment or blindness
Full text linkIndependent mobility is a challenge for people with visual impairment or blindness. Groundbreaking innovation comes from mobile devices (e.g., smartphones) that are convenient platforms to provide assistive technologies in the form of mobile applications. This paper presents ZebraRecognizer, a software module that recognizes zebra crossings and that advances state-of-the-art along two directions. First, it removes projection distortion from the acquired image, hence improving the accuracy of the recognition and making it possible to compute the quantified relative position of the crossing with respect to the user, which is crucial to effectively guide the user. Second, ZebraRecognizer is efficient, as it adopts a customized version of the EDLines algorithm that is also implemented to run in parallel on the GPU. Experimental results show that ZebraRecognizer is accurate, efficient and it computes the crossings position precisely
- …
