A novel deep hybrid framework with dynamic composite fitness optimization for robust real-time human activity recognition
PeerJ Computer Science · 16.11.2026 · 12 · e4168
DOI: 10.7717/peerj-cs.4168 ↗Kurz gefasst
Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 16.11.2026. Verfasst von Wasi Ullah und 6 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Princess Nourah bint Abdulrahman University Researchers Supporting Project, Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
Abstract (Originalsprache)
Real-time Human Activity Recognition (HAR) has wide-ranging applications in areas such as context-aware environments, public safety, assistive technologies, and autonomous monitoring and surveillance systems. However, existing real-time HAR systems face significant challenges, including limited scalability and high computational costs arising from redundant features. To address these issues, the Inception-V3 model was customized with region-based and boundary-aware operations, using average pooling and max pooling, respectively, to enhance region homogeneity, suppress noise, and capture discriminative local features, while improving robustness through down-sampling. Furthermore, to effectively encode motion dynamics, an Attention-Augmented Long Short-Term Memory (AA-LSTM) network was employed to learn temporal dependencies across video frames. Features are extracted from the video dataset and are then optimized through a novel proposed dynamic composite feature selection method called Adaptive Dynamic Fitness Sharing and Attention (ADFSA). This ADFSA mechanism is embedded within a genetic algorithm to select a compact, optimized subset of features by dynamically balancing multiple objectives, accuracy, redundancy reduction, feature uniqueness, and complexity minimization. As a result, the selected subset of diverse and discriminative features enables lightweight machine learning classifiers to achieve accurate and robust HAR in heterogeneous environments. Experimental results demonstrate up to 99.65% accuracy using as few as seven selected features, with improved inference time on the challenging University of Central Florida (UCF)-YouTube dataset, which includes factors such as occlusion, cluttered backgrounds, complex motion dynamics, and poor illumination conditions.
Förderung
- Princess Nourah bint Abdulrahman University Researchers Supporting Project, Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
Zitieren
APA
Ullah, W., Khalid, Y. N., Mazroa, A. A., Al-Khatib, O., Wasfi, H., Hoxha, J., & Khan, S. H. (2026). A novel deep hybrid framework with dynamic composite fitness optimization for robust real-time human activity recognition. PeerJ Computer Science, 12, e4168. https://doi.org/10.7717/peerj-cs.4168
Vancouver
Ullah W, Khalid YN, Mazroa AA, Al-Khatib O, Wasfi H, Hoxha J, et al. A novel deep hybrid framework with dynamic composite fitness optimization for robust real-time human activity recognition. PeerJ Computer Science. 2026;12:e4168. doi:10.7717/peerj-cs.4168
BibTeX
@article{Ullah2026,
author = {Ullah, Wasi and Khalid, Yasir Noman and Mazroa, Alanoud Al and Al-Khatib, Obada and Wasfi, Hassan and Hoxha, Julian and Khan, Saddam Hussain},
title = {A novel deep hybrid framework with dynamic composite fitness optimization for robust real-time human activity recognition},
journal = {PeerJ Computer Science},
year = {2026},
volume = {12},
pages = {e4168},
doi = {10.7717/peerj-cs.4168}
}
EUVIMED