Special Issue "Innovation in Machine Intelligence for Critical Infrastructures"

A special issue of Inventions (ISSN 2411-5134). This special issue belongs to the section "Inventions and innovation in Design, Modeling and Computing Methods".

Deadline for manuscript submissions: closed (31 March 2019)

Special Issue Editors

Guest Editor
Dr. Anastasios Doulamis

Photogrammetry and Computer Vision Lab., National Technical University of Athens, Athens 15773, Greece
Website | E-Mail
Interests: image processing; computer vision; robotic systems; deep machine learning; machine learning; markovian models; signal processing and pattern analysis
Guest Editor
Dr. Tania Stathaki

Reader, Communications and Signal Processing Research Group, Department of Electrical and Electronic Engineering, Imperial College London, Exhibition Road, London SW7 2AZ, United Kingdom
Website | E-Mail
Interests: image fusion; computer vision; remote sensing; urban monitoring
Guest Editor
Dr. Athanasios Voulodimos

National Technical University of Athens, Greece
Website | E-Mail
Interests: machine learning; computational intelligence; computer vision; multimedia analysis; pattern analysis; pervasive computing; intelligent systems

Special Issue Information

Dear Colleagues,

Critical infrastructures are among the cornerstones that support modern daily living through the variety of essential services offered to their end-users. Water and gas utility networks, transportation networks (e.g., airports, rail stations), communication networks, and the smart electric power grid are prominent cases of critical infrastructures. The rapid advancements in heterogeneous sensor development, sensor data acquisition, transmission and processing, and the Internet of Things, have created new possibilities for growth within critical infrastructures, but also new risks in terms of theses infrastructures’ vulnerability versus attacks. For example, the increased penetration of renewable sources, cloud storage, controllable loads, such as plug-in electric vehicles, require new approaches for optimal resource management that ensure security and privacy while yielding system-wide benefits. At the same time, the latest developments in machine learning and intelligent information processing act as an enabler for innovative methods to support dynamic system requirements and complex infrastructure dependencies with evolving characteristics, and ensure the safety and security of such infrastructures from both cyber and physical points of view. This Special Issue will feature key innovations in artificial intelligence, machine learning, signal and information processing put forward to advancing the design, analysis, optimization, operation and protection of critical infrastructures.

Keywords

  • machine learning
  • deep learning
  • intelligent systems
  • sensor signal processing
  • sensor networks topology and design
  • resource optimization
  • critical infrastructure protection
  • safety and security
  • infrastructure resilience
  • surveillance

Published Papers (7 papers)

View options order results:
result details:
Displaying articles 1-7
Export citation of selected articles as:

Research

Open AccessArticle Skeleton-Based Human Action Recognition through Third-Order Tensor Representation and Spatio-Temporal Analysis
Received: 17 December 2018 / Revised: 1 February 2019 / Accepted: 3 February 2019 / Published: 8 February 2019
PDF Full-text (1784 KB) | HTML Full-text | XML Full-text
Abstract
Given the broad range of applications from video surveillance to human–computer interaction, human action learning and recognition analysis based on 3D skeleton data are currently a popular area of research. In this paper, we propose a method for action recognition using depth sensors [...] Read more.
Given the broad range of applications from video surveillance to human–computer interaction, human action learning and recognition analysis based on 3D skeleton data are currently a popular area of research. In this paper, we propose a method for action recognition using depth sensors and representing the skeleton time series sequences as higher-order sparse structure tensors to exploit the dependencies among skeleton joints and to overcome the limitations of methods that use joint coordinates as input signals. To this end, we estimate their decompositions based on randomized subspace iteration that enables the computation of singular values and vectors of large sparse matrices with high accuracy. Specifically, we attempt to extract different feature representations containing spatio-temporal complementary information and extracting the mode-n singular values with regards to the correlations of skeleton joints. Then, the extracted features are combined using discriminant correlation analysis, and a neural network is used to recognize the action patterns. The experimental results presented use three widely used action datasets and confirm the great potential of the proposed action learning and recognition method. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Figure 1

Open AccessArticle A Distance-Dependent Chinese Restaurant Process Based Method for Event Detection on Social Media
Received: 2 November 2018 / Revised: 30 November 2018 / Accepted: 4 December 2018 / Published: 7 December 2018
PDF Full-text (1100 KB) | HTML Full-text | XML Full-text
Abstract
In this paper, we propose a method for event detection on social media, which aims at clustering media items into groups of events based on their textural information as well as available metadata. Our approach is based on distance-dependent Chinese Restaurant Process (ddCRP), [...] Read more.
In this paper, we propose a method for event detection on social media, which aims at clustering media items into groups of events based on their textural information as well as available metadata. Our approach is based on distance-dependent Chinese Restaurant Process (ddCRP), a clustering approach resembling Dirichlet process algorithm. Furthermore, we scrutinize the effectiveness of a series of pre-processing steps in improving the detection performance. We experimentally evaluated our method using the Social Event Detection (SED) dataset of MediaEval 2013 benchmarking workshop, which pertains to the discovery of social events and their grouping in event-specific clusters. The obtained results indicate that the proposed method attains very good performance rates compared to existing approaches. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Figure 1

Open AccessArticle Digitization and Visualization of Folk Dances in Cultural Heritage: A Review
Received: 31 July 2018 / Revised: 8 October 2018 / Accepted: 12 October 2018 / Published: 23 October 2018
PDF Full-text (1606 KB) | HTML Full-text | XML Full-text
Abstract
According to UNESCO, cultural heritage does not only include monuments and collections of objects, but also contains traditions or living expressions inherited from our ancestors and passed to our descendants. Folk dances represent part of cultural heritage and their preservation for the next [...] Read more.
According to UNESCO, cultural heritage does not only include monuments and collections of objects, but also contains traditions or living expressions inherited from our ancestors and passed to our descendants. Folk dances represent part of cultural heritage and their preservation for the next generations appears of major importance. Digitization and visualization of folk dances form an increasingly active research area in computer science. In parallel to the rapidly advancing technologies, new ways for learning folk dances are explored, making the digitization and visualization of assorted folk dances for learning purposes using different equipment possible. Along with challenges and limitations, solutions that can assist the learning process and provide the user with meaningful feedback are proposed. In this paper, an overview of the techniques used for the recording of dance moves is presented. The different ways of visualization and giving the feedback to the user are reviewed as well as ways of performance evaluation. This paper reviews advances in digitization and visualization of folk dances from 2000 to 2018. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Figure 1

Open AccessArticle A Robust Information Life Cycle Management Framework for Securing and Governing Critical Infrastructure Systems
Received: 31 July 2018 / Revised: 30 August 2018 / Accepted: 12 October 2018 / Published: 17 October 2018
PDF Full-text (7644 KB) | HTML Full-text | XML Full-text
Abstract
In modern societies, the rampant growth of the Internet, both on the technological and social level, has created fertile ground for the emergence of new types of risk. On top of that, it enhances pre-existing threats by offering new means for accessing and [...] Read more.
In modern societies, the rampant growth of the Internet, both on the technological and social level, has created fertile ground for the emergence of new types of risk. On top of that, it enhances pre-existing threats by offering new means for accessing and exploiting Critical Infrastructures. As the kinds of potential threats evolve, the security, safety and resilience of these infrastructures must be updated accordingly, both at a prevention, as well as a real-time confrontation level. Our research approaches the security of these infrastructures with a focus on the data and utilization of every possible piece of information that derives from this ecosystem. Such a task is quite daunting, since the quantity of data that requires processing resides in the Big Dataspace. To address this, we introduce a new well-defined Information Life Cycle in order to properly model and optimise the way information flows through a modern security system. This life cycle covers all the possible stages, starting from the collection phase up until the exploitation of information intelligence. That ensures the efficiency of data processing and filtering while increasing both the veracity and validity of the final outcome. In addition, an agile Framework is introduced that is optimised to take full advantage of the Information Life Cycle. As a result, it exploits the generated knowledge taking the correct sequence of actions that will successfully address possible threats. This Framework leverages every possible data source that could provide vital information to Critical Infrastructures by performing analysis and data fusion being able to cope with data variety and variability. At the same time, it orchestrates the pre-existing processes and resources of these infrastructures. Through rigorous testing, it was found that response time against hazards was dramatically decreased. As a result, this Framework is an ideal candidate for strengthening and shielding the infrastructures’ resilience while improving management of the resources used. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Graphical abstract

Open AccessArticle Deep Learning Based Surveillance System for Open Critical Areas
Received: 3 August 2018 / Revised: 3 October 2018 / Accepted: 4 October 2018 / Published: 11 October 2018
PDF Full-text (1088 KB) | HTML Full-text | XML Full-text
Abstract
How to automatically monitor wide critical open areas is a challenge to be addressed. Recent computer vision algorithms can be exploited to avoid the deployment of a large amount of expensive sensors. In this work, we propose our object tracking system which, combined [...] Read more.
How to automatically monitor wide critical open areas is a challenge to be addressed. Recent computer vision algorithms can be exploited to avoid the deployment of a large amount of expensive sensors. In this work, we propose our object tracking system which, combined with our recently developed anomaly detection system. can provide intelligence and protection for critical areas. In this work. we report two case studies: an international pier and a city parking lot. We acquire sequences to evaluate the effectiveness of the approach in challenging conditions. We report quantitative results for object counting, detection, parking analysis, and anomaly detection. Moreover, we report state-of-the-art results for statistical anomaly detection on a public dataset. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Figure 1

Open AccessArticle Protection of Service-Oriented Environments Serving Critical Infrastructures
Received: 26 July 2018 / Revised: 20 August 2018 / Accepted: 23 August 2018 / Published: 30 August 2018
PDF Full-text (2027 KB) | HTML Full-text | XML Full-text
Abstract
The emergence of service-oriented architectures has driven the shift towards a service-oriented paradigm, which has been adopted in several application domains. The advent of cloud computing facilities and recently of edge computing environments has increased the aforementioned paradigm shift towards service provisioning. In [...] Read more.
The emergence of service-oriented architectures has driven the shift towards a service-oriented paradigm, which has been adopted in several application domains. The advent of cloud computing facilities and recently of edge computing environments has increased the aforementioned paradigm shift towards service provisioning. In this context, various “traditional” critical infrastructure components have turned to services, being deployed and managed on top of cloud and edge computing infrastructures. However, the latter poses a specific challenge: the services of the critical infrastructures within and across application verticals/domains (e.g., transportation, health, industrial venues, etc.) need to be continuously available with near-zero downtime. In this context, this paper presents an approach for high-performance monitoring and failure detection of critical infrastructure services that are deployed in virtualized environments. The failure detection framework consists of distributed agents (i.e., monitoring services) to ensure timely collection of monitoring data, while it is enhanced with a voting algorithm to minimize the case of false positives. The goal of the proposed approach is to detect failures in datacenters that support critical infrastructures by targeting both the acquisition of monitoring data in a performant way and the minimization of false positives in terms of potential failure detection. The specific approach is the baseline towards decision making and triggering of actions in runtime to ensure service high availability, given that it provides the required data for decision making on time with high accuracy. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Figure 1

Open AccessArticle Frequency-Domain Joint Motion and Disparity Estimation Using Steerable Filters
Received: 28 November 2017 / Revised: 31 January 2018 / Accepted: 3 February 2018 / Published: 6 February 2018
PDF Full-text (8711 KB) | HTML Full-text | XML Full-text
Abstract
In this paper, the problem of joint disparity and motion estimation from stereo image sequences is formulated in the spatiotemporal frequency domain, and a novel steerable filter-based approach is proposed. Our rationale behind coupling the two problems is that according to experimental evidence [...] Read more.
In this paper, the problem of joint disparity and motion estimation from stereo image sequences is formulated in the spatiotemporal frequency domain, and a novel steerable filter-based approach is proposed. Our rationale behind coupling the two problems is that according to experimental evidence in the literature, the biological visual mechanisms for depth and motion are not independent of each other. Furthermore, our motivation to study the problem in the frequency domain and search for a filter-based solution is based on the fact that, according to early experimental studies, the biological visual mechanisms can be modelled based on frequency-domain or filter-based considerations, for both the perception of depth and the perception of motion. The proposed framework constitutes the first attempt to solve the joint estimation problem through a filter-based solution, based on frequency-domain considerations. Thus, the presented ideas provide a new direction of work and could be the basis for further developments. From an algorithmic point of view, we additionally extend state-of-the-art ideas from the disparity estimation literature to handle the joint disparity-motion estimation problem and formulate an algorithm that is evaluated through a number of experimental results. Comparisons with state-of-the-art-methods demonstrate the accuracy of the proposed approach. Full article
(This article belongs to the Special Issue Innovation in Machine Intelligence for Critical Infrastructures)
Figures

Figure 1

Inventions EISSN 2411-5134 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top