scispace - formally typeset
Search or ask a question
JournalISSN: 2076-3417

Applied Sciences 

Multidisciplinary Digital Publishing Institute
About: Applied Sciences is an academic journal published by Multidisciplinary Digital Publishing Institute. The journal publishes majorly in the area(s): Computer science & Artificial intelligence. It has an ISSN identifier of 2076-3417. It is also open access. Over the lifetime, 50124 publications have been published receiving 360946 citations.


Papers
More filters
Journal ArticleDOI
TL;DR: This paper presents and discusses various metrics proposed for evaluation of polyphonic sound event detection systems used in realistic situations where there are typically multiple sound sources active simultaneously.
Abstract: This paper presents and discusses various metrics proposed for evaluation of polyphonic sound event detection systems used in realistic situations where there are typically multiple sound sources active simultaneously The system output in this case contains overlapping events, marked as multiple sounds detected as being active at the same time The polyphonic system output requires a suitable procedure for evaluation against a reference Metrics from neighboring fields such as speech recognition and speaker diarization can be used, but they need to be partially redefined to deal with the overlapping events We present a review of the most common metrics in the field and the way they are adapted and interpreted in the polyphonic case We discuss segment-based and event-based definitions of each metric and explain the consequences of instance-based and class-based averaging using a case study In parallel, we provide a toolbox containing implementations of presented metrics

493 citations

Journal ArticleDOI
TL;DR: The advantages of high efficiency and long life span of conventional LED chips are inherited by miniaturized ones as the size gets smaller, the resolution enhances, but at the expense of elevating the complexity of fabrication as mentioned in this paper.
Abstract: Displays based on inorganic light-emitting diodes (LED) are considered as the most promising one among the display technologies for the next-generation The chip for LED display bears similar features to those currently in use for general lighting, but it size is shrunk to below 200 microns Thus, the advantages of high efficiency and long life span of conventional LED chips are inherited by miniaturized ones As the size gets smaller, the resolution enhances, but at the expense of elevating the complexity of fabrication In this review, we introduce two sorts of inorganic LED displays, namely relatively large and small varieties The mini-LEDs with chip sizes ranging from 100 to 200 μm have already been commercialized for backlight sources in consumer electronics applications The realized local diming can greatly improve the contrast ratio at relatively low energy consumptions The micro-LEDs with chip size less than 100 μm, still remain in the laboratory The full-color solution, one of the key technologies along with its three main components, red, green, and blue chips, as well color conversion, and optical lens synthesis, are introduced in detail Moreover, this review provides an account for contemporary technologies as well as a clear view of inorganic and miniaturized LED displays for the display community

418 citations

Journal ArticleDOI
TL;DR: A novel deep learning framework for the detection of pneumonia using the concept of transfer learning, where features from images are extracted using different neural network models pretrained on ImageNet, which then are fed into a classifier for prediction.
Abstract: Pneumonia is among the top diseases which cause most of the deaths all over the world. Virus, bacteria and fungi can all cause pneumonia. However, it is difficult to judge the pneumonia just by looking at chest X-rays. The aim of this study is to simplify the pneumonia detection process for experts as well as for novices. We suggest a novel deep learning framework for the detection of pneumonia using the concept of transfer learning. In this approach, features from images are extracted using different neural network models pretrained on ImageNet, which then are fed into a classifier for prediction. We prepared five different models and analyzed their performance. Thereafter, we proposed an ensemble model that combines outputs from all pretrained models, which outperformed individual models, reaching the state-of-the-art performance in pneumonia recognition. Our ensemble model reached an accuracy of 96.4% with a recall of 99.62% on unseen data from the Guangzhou Women and Children’s Medical Center dataset.

417 citations

Journal ArticleDOI
TL;DR: In this article, the authors present a literature review of the current state-of-the-art of virtual inertia implementation techniques and explore potential research directions and challenges, and discuss several research needs, especially for systems level integration of VINs.
Abstract: The modern power system is progressing from a synchronous machine-based system towards an inverter-dominated system, with large-scale penetration of renewable energy sources (RESs) like wind and photovoltaics. RES units today represent a major share of the generation, and the traditional approach of integrating them as grid following units can lead to frequency instability. Many researchers have pointed towards using inverters with virtual inertia control algorithms so that they appear as synchronous generators to the grid, maintaining and enhancing system stability. This paper presents a literature review of the current state-of-the-art of virtual inertia implementation techniques, and explores potential research directions and challenges. The major virtual inertia topologies are compared and classified. Through literature review and simulations of some selected topologies it has been shown that similar inertial response can be achieved by relating the parameters of these topologies through time constants and inertia constants, although the exact frequency dynamics may vary slightly. The suitability of a topology depends on system control architecture and desired level of detail in replication of the dynamics of synchronous generators. A discussion on the challenges and research directions points out several research needs, especially for systems level integration of virtual inertia systems.

416 citations

Journal ArticleDOI
TL;DR: A taxonomy of IDS is proposed that takes data objects as the main dimension to classify and summarize machine learning- based and deep learning-based IDS literature, and believes that this type of taxonomy framework is fit for cyber security researchers.
Abstract: Networks play important roles in modern life, and cyber security has become a vital research area. An intrusion detection system (IDS) which is an important cyber security technique, monitors the state of software and hardware running in the network. Despite decades of development, existing IDSs still face challenges in improving the detection accuracy, reducing the false alarm rate and detecting unknown attacks. To solve the above problems, many researchers have focused on developing IDSs that capitalize on machine learning methods. Machine learning methods can automatically discover the essential differences between normal data and abnormal data with high accuracy. In addition, machine learning methods have strong generalizability, so they are also able to detect unknown attacks. Deep learning is a branch of machine learning, whose performance is remarkable and has become a research hotspot. This survey proposes a taxonomy of IDS that takes data objects as the main dimension to classify and summarize machine learning-based and deep learning-based IDS literature. We believe that this type of taxonomy framework is fit for cyber security researchers. The survey first clarifies the concept and taxonomy of IDSs. Then, the machine learning algorithms frequently used in IDSs, metrics, and benchmark datasets are introduced. Next, combined with the representative literature, we take the proposed taxonomic system as a baseline and explain how to solve key IDS issues with machine learning and deep learning techniques. Finally, challenges and future developments are discussed by reviewing recent representative studies.

413 citations

Performance
Metrics
No. of papers from the Journal in previous years
YearPapers
20237,659
202212,565
202110,625
20209,132
20195,760
20182,784