Explore the vast range of titles available.

We present a novel approach for learning a finite mixture model on a Riemannian manifold in which Euclidean metrics are not applicable and one needs to resort to geodesic distances consistent with the manifold geometry. For this purpose, we draw inspiration on a variant of the expectation-maximization algorithm, that uses a minimum message length criterion to automatically estimate the optimal number of components from multivariate data lying on an Euclidean space. In order to use this approach on Riemannian manifolds, we propose a formulation in which each component is defined on a different tangent space, thus avoiding the problems associated with the loss of accuracy produced when linearizing the manifold with a single tangent space. Our approach can be applied to any type of manifold for which it is possible to estimate its tangent space. Additionally, we consider using shrinkage covariance estimation to improve the robustness of the method, especially when dealing with very sparsely distributed samples. We evaluate the approach on a number of situations, going from data clustering on manifolds to combining pose and kinematics of articulated bodies for 3D human pose tracking. In all cases, we demonstrate remarkable improvement compared to several chosen baselines.

More than 75% of Internet traffic is now encrypted, and this percentage is constantly increasing. The majority of communications are secured using common encryption protocols such as SSL/TLS and IPsec to ensure security and protect the privacy of Internet users. However, encryption can be exploited to hide malicious activities, camouflaged into normal network traffic. Traditionally, network traffic inspection is based on techniques like deep packet inspection (DPI). Common applications for DPI include but are not limited to firewalls, intrusion detection and prevention systems, L7 filtering, and packet forwarding. With the widespread adoption of network encryption though, DPI tools that rely on packet payload content are becoming less effective, demanding the development of more sophisticated techniques in order to adapt to current network encryption trends. In this work, we present HeaderHunter, a fast signature-based intrusion detection system even for encrypted network traffic. We generate signatures using only network packet metadata extracted from packet headers. In addition, we examine the processing acceleration of the intrusion detection engine using different heterogeneous hardware architectures.

We compared general and specialized databases, by searching bibliographic information regarding journal articles in the computer science field, and by evaluating their bibliographic coverage and the quality of the bibliographic records retrieved. We selected a sample of computer science articles from an Italian university repository (AIR) to carry out our comparison. The databases selected were INSPEC, Scopus, Web of Science (WoS), and DBLP. We found that DBLP and Scopus indexed the highest number of unique articles (4.14 and 4.05 % respectively), that each of the four databases indexed a set of unique articles, that 12.95 % of the articles sampled were not indexed in any of the databases selected, that Scopus was better than WoS for identifying computer science publications, and that DBLP had a greater number of unique articles indexed (19.03 %), when compared to INSPEC (11.28 %). We also measured the quality of a set of bibliographic records, by comparing five databases: Scopus, WoS, INSPEC, DBLP and Google Scholar (GS). We found that WoS, INSPEC and Scopus provided better quality indexing and better bibliographic records in terms of accuracy, control and granularity of information, when compared to GS and DBLP. WoS and Scopus also provided more sophisticated tools for measuring trends of scholarly publications.

In this article, an approach used for the inspection tasks in the WRS2020 Plant Disaster Prevention Challenge is explained. The tasks were categorized into three categories: reading pressure gauges, inspecting rust on a tank, and inspecting cracks in a tank. For reading pressure gauges, the “you only look once” algorithm was used to focus on a specific pressure gauge and check the pressure gauge range strings on the gauge using optical character recognition algorithm. Finally, a previously learned classifier was used to read the values shown in the gauge. For rust inspection, image processes were used to focus on a target plate that may be rusted for rust detection. In particular, it was necessary to report the rust area and distribution type. Thus, the pixel ratio and grouping of rust were used to count the rust. The approach for crack inspection was similar to that for rust. The target plate was focused on first, and then the length of the crack was measured using image processing. Its width was not measured but was calculated using the crack area and length. For each system developed to approach each task, the results of the preliminary experiment and those of WRS2020 are shown. Finally, the approaches are summarized, and planned future work is discussed.

This paper presents an advanced methodology for the detection of damage in aircraft composite materials based on the sensor fusion of two image-based non-destructive evaluation techniques. Both of the techniques, phased-array ultrasonics and infra-red thermography, are benchmarked on an aircraft-grade painted composite material skin panel with stringers. The sensors systems for carrying out the inspections have been developed and miniaturized for being integrated on a vortex-robotic platform inspector, in the framework of a larger research initiative, the Horizon-2020 ‘CompInnova’ project.

Based on the data of indexed papers from Inspec database, scientometrics and statistics‐related analysis methods are used to study the overall development trend of international cooperation in advanced manufacturing, the characteristics and evolution of major countries in four dimensions of cooperation scale, cooperation intensity, cooperation dominance rate, cooperation impact, and cooperation network from 2011 to 2020, and this article focuses on the changing trend of China and the differences existing with other countries. The study shows that cooperation in advanced manufacturing has an upward trend, with China and the US having a clear lead in cooperation scale and intensity. China leads the most in the aspect of international cooperation in the field of advanced manufacturing, while the US. is at the heart of a network of cooperation across the field. Based on the data of indexed papers from Inspec database, scientometrics and statistics‐related analysis methods are used to study the overall development trend of international cooperation in advanced manufacturing.

The present study explores the quantitative analysis of literature on gravity from 2015 to 2019 based on INSPEC database. Scientometric is the study of quantitative analysis of scientific documents. For the present study data was collected from the INSPEC database published by Elsevier during the period from 2015 to 2019. The study evaluated the different dimensions of the scientometric analysis like, growth of literature, institution wise distribution, most prolific author, relative growth rate and doubling time etc. Highest number (7562) of articles published in the year 2018. The relating growth rate is decreasing and doubling time is increasing over the period. Majority of the articles are in the form of journals, China is the most productive country in global level, Sharif, M is the most prolific author in the field of gravity for the period of 2015 to 2019.

PurposeAs library budgets continue to constrict, librarians will need to become more familiar with comparing database host platforms. This paper aims to compare Inspec on Elsevier’s Engineering Village (EV) and Clarivate’s Web of Science (WOS) from a novice user experience. The main objectives are to identify some R1 institutions that subscribe to Inspec and highlight some of the key differences between the two platforms.Design/methodology/approachInformation on Inspec was gathered from various sources such as the home website, IET, and the host platform websites of Elsevier and Clarivate Analytics. Additional evidence was also collected from brochures and guides to help illustrate some of the main features and differences that novice users would be familiar with.FindingsMost institutions subscribe to Inspec via the EV platform. Results from the study conclude that EV was the choice over WOS for hosting Inspec owing to a more user-friendly interface, potential lower cost and faster platform updates in response to meeting user needs.Research limitations/implicationsAs database platforms change over time, feature areas such as content, interface and features remain important for information professionals and librarians to stay current with those changes. Also, this work can help librarians with understanding the planning and developing of a process for comparing identical or similar databases on different platforms.Originality/valueMuch of the literature focuses on the unfamiliar details and not so much on the novice user. This paper provides a unique perspective in how a novice user would prefer the attributes of one host platform over the other. Additionally, the same review criteria can be applied in other subjects and disciplines.

Predation is thought to be one of the main structuring forces in animal communities. However, se- lective predation is often measured on isolated traits in response to a single predatory species, but only rarely are selective forces on several traits quantified or even compared between different predators naturally occurring in the same system. In the present study, we therefore measured be- havioral and morphological traits in young-of-the-year Eurasian perch Perca fluviatilis and com- pared their selective values in response to the 2 most common predators, adult perch and pike Esox lucius. Using mixed effects models and model averaging to analyze our data, we quantified and compared the selectivity of the 2 predators on the different morphological and behavioral traits. We found that selection on the behavioral traits was higher than on morphological traits and perch predators preyed overall more selectively than pike predators. Pike tended to positively se- lect shallow bodied and nonvigilant individuals （i.e. individuals not performing predator inspec- tion）. In contrast, perch predators selected mainly for bolder juvenile perch （i.e. individuals spend- ing more time in the open, more active）, which was most important. Our results are to the best of our knowledge the first that analyzed behavioral and morphological adaptations of juvenile perch facing 2 different predation strategies. We found that relative specific predation intensity for the di- vergent traits differed between the predators, providing some additional ideas why juvenile perch display such a high degree of phenotypic plasticity.