A5100443536- Context-Aware Activity Recognition Systems
- Advanced Image and Video Retrieval Techniques
- IoT and Edge/Fog Computing
- Image Retrieval and Classification Techniques
- Technology Use by Older Adults
- Video Analysis and Summarization
- Distributed and Parallel Computing Systems
- Semantic Web and Ontologies
- Music and Audio Processing
- Anomaly Detection Techniques and Applications
- Service-Oriented Architecture and Web Services
- Robotics and Automated Systems
- Scientific Computing and Data Management
- Speech and Audio Processing
- Indoor and Outdoor Localization Technologies
- Network Security and Intrusion Detection
- Human Pose and Action Recognition
- Human Mobility and Location-Based Analysis
- Atmospheric and Environmental Gas Dynamics
- Advanced Neural Network Applications
- Recommender Systems and Techniques
- Atmospheric Ozone and Climate
- Advanced Malware Detection Techniques
- Visual Attention and Saliency Detection
- Multimodal Machine Learning Applications
University of Ulster
2012-2025
Dalian University of Technology
2024-2025
Chongqing University
2004-2025
Deyang Stomatological Hospital
2024-2025
Nanchang University
2010-2024
Xi'an Jiaotong University
2018-2024
State Key Laboratory of Electrical Insulation and Power Equipment
2024
Université Claude Bernard Lyon 1
2009-2024
University of California, Irvine
2024
Laboratoire d'Informatique en Images et Systèmes d'Information
2012-2024
Research on sensor-based activity recognition has, recently, made significant progress and is attracting growing attention in a number of disciplines application domains. However, there lack high-level overview this topic that can inform related communities the research state art. In paper, we present comprehensive survey to examine development current status various aspects recognition. We first discuss general rationale distinctions vision-based Then, review major approaches methods...
The Atmospheric Infrared Sounder (AIRS) and its two companion microwave sounders, AMSU HSB were launched into polar orbit onboard the NASA Aqua Satellite in May 2002. required sounding system to provide high-quality research data for climate studies meet NOAA's requirements improving operational weather forecasting. NOAA requirement translated global retrieval of temperature humidity profiles with accuracies approaching those radiosondes. AIRS also provides new measurements several...
Carbamazepine, an anticonvulsant and a mood-stabilizing drug, is the main cause of Stevens-Johnson syndrome (SJS) its related disease, toxic epidermal necrolysis (TEN), in Southeast Asian countries. Carbamazepine-induced SJS-TEN strongly associated with HLA-B*1502 allele. We sought to prevent carbamazepine-induced by using screening prospectively identify subjects at genetic risk for condition.From 23 hospitals Taiwan, we recruited 4877 candidate who had not taken carbamazepine. genotyped...
This paper introduces a knowledge-driven approach to real-time, continuous activity recognition based on multisensor data streams in smart homes. The goes beyond the traditional data-centric methods for three ways. First, it makes extensive use of domain knowledge life cycle recognition. Second, uses ontologies explicit context and modeling representation. Third finally, exploits semantic reasoning classification inferencing, thus enabling both coarse-grained fine-grained In this paper, we...
N-version programing is defined as the independent generation of N2 2 functionally e quivalent prograns fran sme initial specification. A methodology has been devised and three types special mechanisms have identified t hat are needed to coordinate execution an software unit canpare correspondent results generated by each version. Two experiments conducted test feasibility programing. The o f these discussed. In addition, constraints that must be met for effective application
Vehicular social networking is an emerging application of the Internet Vehicles (IoV) which aims to achieve seamless integration vehicular networks and networks. However, unique characteristics networks, such as high mobility frequent communication interruptions, make content delivery end-users under strict delay constraints extremely challenging. In this paper, we propose a social-aware edge computing architecture that solves problem by using some vehicles in network servers can store...
Abstract Chemical cleaning and disinfection are crucial steps for eliminating infection in root canal treatment. However, irrigant selection or irrigation procedures far from clear. The vapor lock effect the apical region has yet to be solved, impeding efficacy resulting residual infections compromised treatment outcomes. Additionally, ambiguous clinical indications medication non-standardized dressing protocols must clarified. Inappropriate intracanal may present side effects jeopardize...
Purpose This paper aims to serve two main purposes. In the first instance it provide an overview addressing state‐of‐the‐art in area of activity recognition, particular, object‐based recognition. will necessary material inform relevant research communities latest developments this addition providing a reference for researchers and system developers who ware working towards design development activity‐based context aware applications. second introduces novel approach recognition based on use...
Human activity has increased the concentration of earth's atmospheric carbon dioxide, which plays a direct role in contributing to global warming. Mid‐tropospheric CO 2 retrieved by Atmospheric Infrared Sounder shows substantial spatiotemporal variability that is supported situ aircraft measurements. The distribution middle tropospheric strongly influenced surface sources and large‐scale circulations such as mid‐latitude jet streams synoptic weather systems, most notably summer hemisphere....
Deep CNN-based object detection systems have achieved remarkable success on several large-scale benchmarks. However, training such detectors requires a large number of labeled bounding boxes, which are more difficult to obtain than image-level annotations. Previous work addresses this issue by transforming classifiers into detectors. This is done modeling the differences between two categories with both imagelevel and box annotations, transferring information convert for without We improve...
The global population is aging; projections show that by 2050, more than 20% of the will be aged over 64. This lead to an increase in aging related illness, a decrease informal support, and ultimately issues with providing care for these individuals. Assistive smart homes provide promising solution some issues. Nevertheless, they currently have hindering their adoption. To help address issues, this study introduces novel approach implementing assistive homes. devised based upon intention...
Two low molecular weight gelators containing 4-ethynyl-1,8-naphthalimide groups with large conjugated structure via different length of alkyl chains were synthesized and fully characterized. The gelation properties, structural character, fluorescence the gels investigated methods scanning electron microscopy, X-ray diffraction, spectral studies. have high quantum yields in both solution solid state. Interestingly, wavelength fluorescent emission reversible sol-gel transition process has a...
Activity models play a critical role for activity recognition and assistance in ambient assisted living. Existing approaches to modeling suffer from number of problems, e.g., cold-start, model reusability, incompleteness. In an effort address these we introduce ontology-based hybrid approach that combines domain knowledge based specification data-driven learning. Central the is iterative process begins with "seed" created by ontological engineering. The are deployed, subsequently evolved...
Social Network Analysis is attracting growing attention as social networking sites and their enabled applications transform impact society. This paper aims to provide a comprehensive review of network analysis state the art research practice. In authors’ first examine core concepts ingredients analysis. Secondly, they trend related research. The then consider modelling motivations, discussing models in line with tie formation approaches, where connections between nodes are taken into...
Abstract The use of Convolutional Neural Networks (CNNs) as a feature learning method for Human Activity Recognition (HAR) is becoming more and common. Unlike conventional machine methods, which require domain-specific expertise, CNNs can extract features automatically. On the other hand, training phase, making them prone to cold-start problem. In this work, case study presented where pre-trained CNN extractor evaluated under realistic conditions. consists two main steps: (1) different...