Transfer Learning-Based Method for Classifying Yoga Poses Using Deep Convolutional Neural Networks DOI

Nagalakshmi Vallabhaneni,

Prabhavathy Panneer,

M. Venkatesan

et al.

2019 Innovations in Power and Advanced Computing Technologies (i-PACT), Journal Year: 2023, Volume and Issue: unknown, P. 1 - 6

Published: Dec. 8, 2023

Yoga is a popular practice that aims to improve one's health and well-being through physical postures, breathing exercises, meditation. The growing popularity of yoga has prompted researchers focus on automating the process classifying poses. Using deep convolutional neural networks (CNNs) transfer learning, we present learning-based method for correctly categorizing postures in this research. primary goal study develop an efficient automatically recognizing poses given photographs. impressive success CNNs picture classification tasks exploited here accomplish goal. We also use customizing pre-trained CNN models our unique posture categorization task. This paper presents Transfer Learning based Method Poses using Deep Convolutional Neural Networks (TLMYPC-DCNN). proposed model uses Posture Dataset images. Initially, starts with pre-processing input images by splitting data into three groups such as training set, validation set test data. learning will be trained data, then its parameters tweaked Finally, used assess how well performs. A network (CNN) called MobileNetV2 applied photos. experimental results show suggested TLMYPC-DCNN technique effective at accurately reliably

Language: Английский

YoNet: A Neural Network for Yoga Pose Classification DOI Creative Commons
Faisal Bin Ashraf, Muhammad Usama Islam,

Md. Rayhan Kabir

et al.

SN Computer Science, Journal Year: 2023, Volume and Issue: 4(2)

Published: Feb. 8, 2023

Yoga has become an integral part of human life to maintain a healthy body and mind in recent times. With the growing, fast-paced work from home, it difficult for people invest time gymnasium exercises. Instead, they like do assisted exercises at home where pose recognition techniques play most vital role. Recognition different poses is challenging due proper dataset classification architecture. In this work, we have proposed deep learning-based model identify five yoga comparatively fewer amounts data. We compared our model's performance with some state-of-the-art image models-ResNet, InceptionNet, InceptionResNet, Xception found architecture superior. Our extracts spatial, depth features individually considers them further calculation classification. The experimental results show that achieved 94.91% accuracy 95.61% precision.

Language: Английский

Citations

37

Yoga Meets Intelligent Internet of Things: Recent Challenges and Future Directions DOI Creative Commons

Rishi Pal,

Deepak Adhikari, Md Belal Bin Heyat

et al.

Bioengineering, Journal Year: 2023, Volume and Issue: 10(4), P. 459 - 459

Published: April 9, 2023

The physical and mental health of people can be enhanced through yoga, an excellent form exercise. As part the breathing procedure, yoga involves stretching body organs. guidance monitoring are crucial to ripe full benefits it, as wrong postures possess multiple antagonistic effects, including hazards stroke. detection possible with Intelligent Internet Things (IIoT), which is integration intelligent approaches (machine learning) (IoT). Considering increment in practitioners recent years, IIoT has led successful implementation IIoT-based training systems. This paper provides a comprehensive survey on integrating IIoT. also discusses types procedure for using Additionally, this highlights various applications safety measures, challenges, future directions. latest developments findings its

Language: Английский

Citations

32

IoMT Meets Machine Learning: From Edge to Cloud Chronic Diseases Diagnosis System DOI Creative Commons
Natasha Nigar, Abdul Jaleel, Shahid Islam

et al.

Journal of Healthcare Engineering, Journal Year: 2023, Volume and Issue: 2023, P. 1 - 13

Published: Jan. 1, 2023

In conventional healthcare, real-time monitoring of patient records and information mining for timely diagnosis chronic diseases under certain health conditions is a crucial process. Chronic diseases, if not diagnosed in time, may result patients’ death. modern medical healthcare systems, Internet Things (IoT) driven ecosystems use autonomous sensors to sense track suggest appropriate actions. this paper, novel IoT machine learning (ML)-based hybrid approach proposed that considers multiple perspectives early detection 6 different such as COVID-19, pneumonia, diabetes, heart disease, brain tumor, Alzheimer’s. The results from ML models are compared accuracy, precision, recall, F1 score, area the curve (AUC) performance measure. validated cloud-based environment using benchmark real-world datasets. statistical analyses on datasets ANOVA tests show accuracy classifiers significantly different. This will help sector doctors diseases.

Language: Английский

Citations

13

EfficientNets transfer learning strategies for histopathological breast cancer image analysis DOI
Sakinat Oluwabukonla Folorunso,

Joseph Bamidele Awotunde,

Yagateela Pandu Rangaiah

et al.

Advances in Complex Systems, Journal Year: 2023, Volume and Issue: 15(02)

Published: April 5, 2023

Breast cancer (BC) is one of the major principal sources high mortality among women worldwide. Consequently, early detection essential to save lives. BC can be diagnosed with different modes medical images such as mammography, ultrasound, computerized tomography, biopsy, and magnetic resonance imaging. A histopathology study (biopsy) that results in often performed help diagnose analyze BC. Transfer learning (TL) a machine-learning (ML) technique reuses method initially built for task applied model new task. TL aims enhance assessment desired learners by moving knowledge contained another but similar source domain. challenge small dataset domain reduced build learners. plays role image analysis because this immense property. This paper focuses on use methods investigation classification detection, preprocessing, pretrained models, ML models. Through empirical experiments, EfficientNets neural network architectures models were built. The support vector machine eXtreme Gradient Boosting (XGBoost) learned dataset. result showed comparative good performance EfficientNetB4 XGBoost. An outcome based accuracy, recall, precision, F1_Score XGBoost 84%, 0.80, 0.83, 0.81, respectively.

Language: Английский

Citations

10

Thermal infrared imaging based on wearable sensing devices for sports athlete body posture detection: Human thermal energy regulation DOI
Mei Wang,

Liyu Liang

Thermal Science and Engineering Progress, Journal Year: 2025, Volume and Issue: unknown, P. 103282 - 103282

Published: Jan. 1, 2025

Language: Английский

Citations

0

Enhancing Human Posture Detection with Hybrid Deep VGG16 and Attention Mechanism Network DOI Open Access
Roseline Oluwaseun Ogundokun, Rytis Maskeliūnas,

Federick Oscar

et al.

Procedia Computer Science, Journal Year: 2025, Volume and Issue: 258, P. 1209 - 1218

Published: Jan. 1, 2025

Language: Английский

Citations

0

Application of Human Posture Recognition and Classification in Performing Arts Education DOI Creative Commons
Jing Shen, Ling Chen

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 125906 - 125919

Published: Jan. 1, 2024

Language: Английский

Citations

3

ERG-AI: enhancing occupational ergonomics with uncertainty-aware ML and LLM feedback DOI Creative Commons
Sagar Sen, Víctor González, Erik Johannes Husom

et al.

Applied Intelligence, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 10, 2024

Language: Английский

Citations

3

Integrating Deep Learning and Energy Management Standards for Enhanced Solar–Hydrogen Systems: A Study Using MobileNetV2, InceptionV3, and ISO 50001:2018 DOI Creative Commons
Salaki Reynaldo Joshua,

Yang Junghyun,

Sanguk Park

et al.

Hydrogen, Journal Year: 2024, Volume and Issue: 5(4), P. 819 - 850

Published: Nov. 10, 2024

This study addresses the growing need for effective energy management solutions in university settings, with particular emphasis on solar–hydrogen systems. The study’s purpose is to explore integration of deep learning models, specifically MobileNetV2 and InceptionV3, enhancing fault detection capabilities AIoT-based environments, while also customizing ISO 50001:2018 standards align unique needs academic institutions. Our research employs comparative analysis two models terms their performance detecting solar panel defects assessing accuracy, loss values, computational efficiency. findings reveal that achieves 80% making it suitable resource-constrained InceptionV3 demonstrates superior accuracy 90% but requires more resources. concludes both offer distinct advantages based application scenarios, emphasizing importance balancing efficiency when selecting appropriate system management. highlights critical role continuous improvement leadership commitment successful implementation universities.

Language: Английский

Citations

2

DeepPose: An Integrated Deep Learning Model for Posture Detection Using Image and Skeletal Data DOI

Manvendra Singh,

Md. Sarfaraj Alam Ansari, Mahesh Chandra Govil

et al.

2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT), Journal Year: 2023, Volume and Issue: unknown

Published: July 6, 2023

Identifying human actions and postures presents significant challenges for computerized systems. The categorization of these tasks holds particular relevance in the fields health robotics. Leveraging artificial intelligence technologies, it becomes feasible to define classify recurring physical movements accurately. Proper posture is further essential rehabilitation patients because affects effectiveness exercise training. Unfortunately, fail follow correct sequence when performing exercises. To pursue problem, a new method proposed recognition pose estimation that does not require wearable devices. model utilizes 2D coordinates derived from poses as inputs with 18 joints body key points, along an image dataset, accurately various postures. This study involved training custom CNN named DeepPose using both keypoint datasets conducting comparative analysis performance two other pre-trained models. result shows dataset outperforms over dataset.

Language: Английский

Citations

5