A Comprehensive Survey on Deep Learning Methods in Human Activity Recognition

Machine Learning and Knowledge Extraction, Год журнала: 2024, Номер 6(2), С. 842 - 876

Опубликована: Апрель 18, 2024

Human activity recognition (HAR) remains an essential field of research with increasing real-world applications ranging from healthcare to industrial environments. As the volume publications in this domain continues grow, staying abreast most pertinent and innovative methodologies can be challenging. This survey provides a comprehensive overview state-of-the-art methods employed HAR, embracing both classical machine learning techniques their recent advancements. We investigate plethora approaches that leverage diverse input modalities including, but not limited to, accelerometer data, video sequences, audio signals. Recognizing challenge navigating vast ever-growing HAR literature, we introduce novel methodology employs large language models efficiently filter pinpoint relevant academic papers. only reduces manual effort also ensures inclusion influential works. provide taxonomy examined literature enable scholars have rapid organized access when studying approaches. Through survey, aim inform researchers practitioners holistic understanding current landscape, its evolution, promising avenues for future exploration.

Язык: Английский

---

Future Outdoor Safety Monitoring: Integrating Human Activity Recognition with the Internet of Physical–Virtual Things

Applied Sciences, Год журнала: 2025, Номер 15(7), С. 3434 - 3434

Опубликована: Март 21, 2025

The convergence of the Internet Physical–Virtual Things (IoPVT) and Metaverse presents a transformative opportunity for safety health monitoring in outdoor environments. This concept paper explores how integrating human activity recognition (HAR) with IoPVT within can revolutionize public safety, particularly urban settings challenging climates architectures. By seamlessly blending physical sensor networks immersive virtual environments, highlights future where real-time data collection, digital twin modeling, advanced analytics, predictive planning proactively enhance well-being. Specifically, three dimensions humans, technology, environment interact toward measuring health, climate. Three cultural scenarios showcase to utilize HAR–IoPVT sensors external staircases, rural climate, coastal infrastructure. Advanced algorithms analytics would identify potential hazards, enabling timely interventions reducing accidents. also societal benefits, such as proactive monitoring, enhanced emergency response, contributions smart city initiatives. Additionally, we address challenges research directions necessary realize this future, emphasizing AI technical scalability, ethical considerations, importance interdisciplinary collaboration designs policies. articulating an AI-driven HAR vision along required advancements edge-based fusion, responsiveness fog computing, social through cloud aim inspire academic community, industry stakeholders, policymakers collaborate shaping technology profoundly improves enhances enriches quality life.

Язык: Английский

---

Device Position-Independent Human Activity Recognition with Wearable Sensors Using Deep Neural Networks

Applied Sciences, Год журнала: 2024, Номер 14(5), С. 2107 - 2107

Опубликована: Март 3, 2024

Human activity recognition (HAR) identifies people’s motions and actions in daily life. HAR research has grown with the popularity of internet-connected, wearable sensors that capture human movement data to detect activities. Recent deep learning advances have enabled more applications using from devices. However, prior often focused on a few sensor locations body. Recognizing real-world activities poses challenges when device positioning is uncontrolled or initial user training are unavailable. This analyzes feasibility models for both position-dependent position-independent HAR. We introduce an advanced residual model called Att-ResBiGRU, which excels at accurate delivers excellent performance evaluate this three public datasets: Opportunity, PAMAP2, REALWORLD16. Comparisons made previously published architectures addressing challenges. The proposed Att-ResBiGRU outperforms existing techniques accuracy, cross-entropy loss, F1-score across all datasets. assess k-fold cross-validation. achieves F1-scores 86.69%, 96.23%, 96.44% REALWORLD16, Opportunity datasets, surpassing state-of-the-art Our experiments analysis demonstrate exceptional applications.

Язык: Английский

---

Human Action Recognition in Smart Living Services and Applications: Context Awareness, Data Availability, Personalization, and Privacy

Опубликована: Июнь 9, 2023

Smart Living, an increasingly prominent concept, entails incorporating sophisticated technologies in homes and urban environments to elevate the quality of life for citizens. A critical success factor Living services applications, from energy management healthcare transportation, is efficacy human action recognition (HAR). HAR, rooted computer vision, seeks identify actions activities using visual data various sensor modalities. This paper extensively reviews literature on HAR amalgamating key contributions challenges while providing insights into future research directions. The review delves essential aspects state art potential societal implications this technology. Moreover, meticulously examines primary application sectors that stand gain such as smart homes, healthcare, cities. By underscoring significance four dimensions Context Awareness, Data Availability, Personalization, Privacy serves a valuable resource researchers practitioners striving advance applications.

Язык: Английский

---

A Deep Learning-Based Framework Oriented to Pathological Gait Recognition with Inertial Sensors

Sensors, Год журнала: 2025, Номер 25(1), С. 260 - 260

Опубликована: Янв. 5, 2025

Abnormal locomotor patterns may occur in case of either motor damages or neurological conditions, thus potentially jeopardizing an individual’s safety. Pathological gait recognition (PGR) is a research field that aims to discriminate among different walking patterns. A PGR-oriented system benefit from the simulation disorders by healthy subjects, since acquisition actual pathological gaits would require higher experimental time larger sample size. Only few works have exploited abnormal patterns, emulated unimpaired individuals, perform PGR with Deep Learning-based models. In this article, authors present workflow based on convolutional neural networks recognize normal and behaviors means inertial data related nineteen subjects. Although preliminary feasibility study, its promising performance terms accuracy computational pave way for more realistic validation data. light this, classification outcomes could support clinicians early detection tracking rehabilitation advances real time.

Язык: Английский

---

A Comparison of Artificial Intelligence and Human Observation in the Assessment of Cattle Handling and Slaughter

Animals, Год журнала: 2025, Номер 15(9), С. 1325 - 1325

Опубликована: Май 3, 2025

Slaughter facilities use a variety of tools to evaluate animal handling, including but not limited live audits, the remote video auditing, and some AI technologies. The objective this study was determine similarity between human evaluator assessments critical cattle handling outcomes in slaughter plant. One hundred twelve clips stunning from plant United Kingdom were collected. identified presence or absence of: Stunning, Electric Prod Usage, Falling, Pen Crowding, Questionable Handling Events. Three evaluators scored videos for these outcomes. Four different datasets generated, Jaccard indices generated. There high (JI > 0.90) Falls AI. consistency 0.80) Crowding. differences ≥ 0.50) humans when identifying Animal Events adept at events further review. implementation assist with facility environment could be an added tool enhance welfare programs.

Язык: Английский

---

A Multi-Agent System for Service Provisioning in an Internet-of-Things Smart Space Based on User Preferences

Sensors, Год журнала: 2024, Номер 24(6), С. 1764 - 1764

Опубликована: Март 8, 2024

The integration of the Internet Things (IoT) and artificial intelligence (AI) is critical to advancement ambient (AmI), as it enables systems understand contextual information react accordingly. While many solutions focus on user-centric services that provide enhanced comfort support, few expand scenarios in which multiple users are present simultaneously, leaving a significant gap service provisioning. To address this problem, paper presents multi-agent system software agents, aware context, advocate for their users’ preferences negotiate settings achieve satisfy everyone, taking into account flexibility. proposed negotiation algorithm illustrated through smart lighting use case, results analyzed terms concrete defined by user selected resulting from regard

Язык: Английский

---

Generisch-Net: A Generic Deep Model for Analyzing Human Motion with Wearable Sensors in the Internet of Health Things

Sensors, Год журнала: 2024, Номер 24(19), С. 6167 - 6167

Опубликована: Сен. 24, 2024

The Internet of Health Things (IoHT) is a broader version the Things. main goal to intervene autonomously from geographically diverse regions and provide low-cost preventative or active healthcare treatments. Smart wearable IMUs for human motion analysis have proven valuable insights into person’s psychological state, activities daily living, identification/re-identification through gait signatures, etc. existing literature, however, focuses on specificity i.e., problem-specific deep models. This work presents generic BiGRU-CNN model that can predict emotional state person, classify re-identify person in closed-loop scenario. For training validation, we employed publicly available closed-access datasets. data were collected with inertial measurement units mounted non-invasively bodies subjects. Our findings demonstrate achieves an impressive accuracy 96.97% classifying living. Additionally, it re-identifies individuals scenarios 93.71% estimates states 78.20%. study represents significant effort towards developing versatile deep-learning using IMUs, demonstrating promising results across multiple applications.

Язык: Английский

---

Long-Short Term Memory decision model for device argumentation in ambient assisted living

Journal of Ambient Intelligence and Humanized Computing, Год журнала: 2024, Номер 15(12), С. 4073 - 4083

Опубликована: Окт. 12, 2024

Язык: Английский

---